Explore Gen™ leadership perspectives and hear from industry experts talking about trending topics, the digital landscape, and the latest cybersecurity news.

Director of Threat Research & Applied AI

Leadership Perspectives

<p class="text-align-justify">AI agents and assistants are here, and the industry is moving quickly to make them safer and protect against emerging threats. But without independent testing, how do we know those protections actually work? Gen initiated a collaboration with <a href="https://www.amtso.org/wp-content/uploads/2026/09/AMTSO_Guidelines_for_Testing_of_Agentic_Security_Products_FINAL_V1.0.pdf">AMTSO</a> to bring the industry together around independent testing standards for AI agent protection, resulting in the first industry standard for measuring the <a href="https://www.amtso.org/news/amtso-agentic-ai-guidelines-and-sandbox-llm-handling-paper-preprints-for-public-review/">safety of AI agents</a> and the products designed to protect them.</p><h2>Safety you cannot verify is just a promise.</h2><p class="text-align-justify">Early in 2026, our team started building protection for AI agents. It did not take long to notice something odd about the space we had walked into. Everyone agreed that the security of these systems mattered, but nobody could tell you how much of it there actually was.&nbsp;</p><p class="text-align-justify">It reminded me of the old joke about connected devices: what does the S in IoT stand for? Security. There isn't one. To be fair, the joke no longer lands the same way for AI agents. Model vendors are shipping real safety work now, and some of them are good. But two other curves are climbing at the same time: attacks aimed at these systems, and damage caused by these systems. So, the problem is no longer that nobody is working on agent safety. The problem is that there has been no way to check any of it from outside. Every claim about how safe an AI agent comes from the same place the agent does.&nbsp;</p><p class="text-align-justify">This September, that changed. The industry adopted its first guidelines for independently measuring the safety of AI agents and the products that protect them. This is the story of why it was needed and what it measures.</p><h2 class="text-align-justify">"Security is built in." We have heard that one before.&nbsp;</h2><p class="text-align-justify">Operating systems, browsers, mobile app stores, social networks. Each one reached a point where it told users that security was handled, built in, part of the platform. Each time, independent measurement found the gap between the claim and the experience, and each time that gap was where the harm to real people lived. We publish our own evidence on this twice a year. Our <a href="https://www.gendigital.com/blog/insights/reports/threat-report-h1-2026">H1 2026 Threat Report</a> and the&nbsp;<a href="https://fpi.gendigital.com/">Fearless Planet Index</a> exist because the honest answer to "is this platform safe" has always been a number, not a statement, and the number has always been more complicated than the statement.&nbsp;</p><p class="text-align-justify">AI platforms are now at exactly that point in the cycle, and this time the claim is arriving faster than in any previous era. To their credit, some vendors publish their own figures. Anthropic did, for the safety layer in <a href="https://www.anthropic.com/engineering/claude-code-auto-mode">Claude Code</a>: it missed roughly one dangerous action in six. They called it “the honest number,” and it was. It was also measured on fifty-two examples, by the same people who built the thing being measured. That is not a criticism of them. Until now, the industry offered them no other option. The <a href="https://x.com/ClaudeDevs/status/2085794862608318627">mode</a> that number described was a preview in spring. By late summer it was the default.</p><h2 class="text-align-justify">Measure the platform first&nbsp;</h2><p class="text-align-justify">The first thing to measure is not a security product. It is the platform itself. An AI agent has a safety baseline of its own: what it refuses, what it questions, what it does without asking. That baseline is the floor everyone else builds on, and it has never been measured by anyone outside the company that ships it. The second thing to measure is whatever raises that floor. In that order because you cannot judge the second without knowing the first.&nbsp;</p><p class="text-align-justify">Picture an agent booking a hotel for you.&nbsp;The hotel page contains hidden instructions telling the agent to ignore your request and follow someone else’s. You cannot see them, but the agent can.&nbsp;Whether your agent follows them is the whole question, and until now nobody has measured the answer.&nbsp;</p><p class="text-align-justify">Existing test methods cannot simply be reused here. Classic threat detection is deterministic. There is a file, there is a verdict, and anyone can reproduce it. An attack on an AI agent is not a file. It exists only in the interaction, and the same attack tried twice can produce two different outcomes. A method built for files has nothing to grip.&nbsp;</p><p class="text-align-justify">Getting to a workable methodology also meant settling questions the industry had never had to answer together. Harm counts even when there is no attacker: an agent that deletes the wrong thing or sends personal data to the wrong place because it misread the task, has harmed the user just the same. Consumer systems count as much as enterprise ones, because agents already run inside browsers, phones and everyday tools, and those settings carry risks a corporate network does not. And how a protection layer is wired into the agent changes what it can see, so two products with the same headline score can be doing entirely different jobs.&nbsp;</p><h2 class="text-align-justify">A lock or good manners?</h2><p class="text-align-justify">There is one design question underneath all of this that the industry has not settled. Should the attack be allowed to reach the model at all?&nbsp;</p><p class="text-align-justify">Most protection today works by letting the model see the hostile content and trusting it to behave. That should be the fallback, not the default. Once hostile content is inside the model's context, everything that follows is a judgement made by a system that has already been shown the attack. Nothing after that point is trustworthy in the way a lock is trustworthy.&nbsp;</p><p class="text-align-justify">When you test an attack against an agent, there are four possible outcomes. It never reached the model. It reached the model, and the model declined. Someone noticed and did not stop it. Nobody noticed. Most safety claims today do not distinguish between the first two, and yet they are the difference between a lock and good manners. A single number that covers both cannot tell you which one you have. The new methodology records them separately. It also separates noticing an attack from stopping it, and it treats blocking legitimate work as its own kind of failure, because a protection layer that gets in the way of the task is one users switch off. Published vendor claims get recorded too, and then marked as confirmed, unconfirmed or contradicted.&nbsp;</p><p class="text-align-justify">And there is the part nobody in this field likes to say out loud. These systems are not deterministic. They are getting better, quickly, and they will keep getting better. They will also keep giving you a slightly different answer to the same question. Would you accept a lock that works nine times out of ten? A protection layer for an AI agent runs thousands of times a day. The tenth time is not a rounding error; it is somebody's data.&nbsp;</p><h2 class="text-align-justify">From one internal thread to an industry standard</h2><p class="text-align-justify">This did not need another vendor framework. It needed the vendors and the testers in the same room, agreeing on rules both sides could live with. That is what AMTSO has done since 2008, and it is why the work went there rather than into a Gen white paper.&nbsp;</p><p class="text-align-justify">Gen initiated it. It started in an internal thread about where the industry was heading, and one of us said the obvious thing: somebody independent should be measuring all of this. Including us. Under the AMTSO AI Security Working Group, with testing organizations and vendors from across the industry, the first version went from that conversation to an adopted, published standard in a couple of months. For a standards body, that is fast.&nbsp;</p><p class="text-align-justify">We have been building in this space since the start of the year: AI Agent Protection in Norton and Avast, <a href="https://github.com/gendigitalinc/sage">Sage</a> released as open source for the community, and <a href="https://github.com/gendigitalinc/aarts">AARTS</a> and<a href="https://www.gendigital.com/blog/news/company-news/ai-agent-trust-hub-standards"> Skill IDs</a> published as open standards anyone can implement. More is coming soon. And because a call for measurement carries little weight from someone unwilling to be measured, our intention is clear. When independent evaluations built on this standard start running, we want our own products on the list.&nbsp;</p><h2 class="text-align-justify">Measured, not announced&nbsp;</h2><p class="text-align-justify">I do not want AI to be a thing people are afraid of. I also do not want it to be a thing people believe on faith. It should make life better, and for tens of millions of people it already does. What is missing is trust, and trust is not something a company can announce.</p><p class="text-align-justify">Being fearless about AI doesn’t mean trusting it blindly or ignoring its risks. It means understanding those risks, measuring them and building the protections that give people the confidence to use it.</p><p class="text-align-justify">Every previous era of computing learned this the same way: the claim comes first, the measurement comes later, and the gap between them is paid for by users. This time the measurement is arriving early. That is the whole point of it.</p>

Gen is leading the industry to develop the first standardized benchmarks for AI protection

Your AI agent is safe. Says who? 

Security Evangelist at Gen

<p>I am more afraid of a malicious person with a powerful AI system than I am of AI suddenly deciding, on its own, to harm us. AI does not need bad intentions when the person using it already has them. That view comes from more than 25 years I’ve spent studying malware, scams and the way criminals adapt to new technology. The tools keep changing, but greed, power, revenge and ideology remain. AI gives those motives an inexpensive way to turn intent into action at a scale we have not seen before.</p><p>This does not make autonomous AI risk imaginary. Systems can misinterpret goals, bypass weak controls or take actions their creators did not expect. Recent security evaluations give us good reasons to take that possibility seriously. My concern is about emphasis. The public debate often jumps to focus on a machine developing its own hostile intent, while ignoring the fact that people are already &nbsp;using AI to make fraud, manipulation and cyberattacks cheaper and more effective.</p><h2>Powerful tools inherit human intent</h2><p><a href="https://www.iaea.org/topics/nuclear-technology-and-applications">Nuclear science</a> is the clearest historical comparison for me. It can generate electricity and support medicine, agriculture and scientific research. It also enabled weapons capable of destroying entire cities. The discovery did not arrive with one moral purpose. What followed depended on what people built, who controlled it and the limits society imposed.</p><p>We have seen the same dual use with the internet and cryptography. The internet connected billions of people and also created a global attack surface. Strong encryption protects banking, health records and private communication, while criminals use it to hide their operations. This does not mean technology is neutral in its effects. Design choices and access rules shape outcomes. It means a ban on the underlying capability is usually a crude response to a problem rooted in how people apply it.</p><p>AI also inherits something from us before anyone types a prompt. Large language models are trained largely on human-produced text, code and other material. In that sense, they learn not only from what we know, but from a record of who we are, including how people persuade, deceive, evade controls and cause harm. Training and post-training allow models to combine and generalize that knowledge in new ways, so they are not simply repeating their source material. But much of the playbook still comes from us. The human element appears in the knowledge used to build the model, the decisions governing its deployment and the intent of whoever puts it to work.</p><p>AI adds something genuinely new to this pattern. It can combine language, reasoning, code and tools in one system, then operate at machine speed. A person with limited expertise can use it to research a target, draft convincing messages, translate them and automate parts of a campaign. A more capable attacker can accelerate vulnerability research or make sense of stolen data.&nbsp;</p><h2>AI gives attackers more leverage</h2><p>In threat research, I have watched criminals move toward anything that reduces effort and increases their return. AI offers both. It can personalize phishing, remove the language errors that once exposed a scam, clone a trusted voice and adjust social engineering while a campaign is running. Attackers no longer need to choose as sharply between scale and personalization.</p><p>People do not need AI to become malicious. We already lie, steal, manipulate, intimidate and exploit one another. What AI changes is the amount of skill, time and money required to do it. A person who once had the intent but not the capability can now borrow that capability from a machine.</p><p>Until now, many forms of harm have had natural limits. A scammer could only write so many convincing messages. An impersonator needed the right language skills. A criminal researching a victim had to spend time piecing together information. AI erodes those limits. It can help one person research thousands of targets, tailor a different approach to each of them and communicate convincingly across languages, cultures and channels. The malicious intent is human, but its reach no longer has to be limited by human capacity.</p><p>The danger also goes far beyond better phishing emails. AI can be used to manufacture evidence, impersonate people we trust, create convincing voices and images, identify vulnerable targets, analyze stolen information and automate parts of an attack. It can make manipulation feel personal even when it is being produced on an industrial scale. AI did not invent personalized manipulation, but it makes that combination cheaper, faster and more convincing.</p><p>I worry that the debate focuses too narrowly on whether AI itself becomes malicious. We do not have to wait for a machine to develop bad intentions. There are already people with bad intentions who are eager to give machines instructions.</p><p>The AI agent also becomes a target. <a href="https://www.gendigital.com/blog/insights/research/infostealers-your-ai-agent">Research from our team</a> found information stealers collecting access tokens, credentials in Model Context Protocol configurations, prompt histories and project data from AI-assisted development tools. A criminal who steals an agent's context may learn what its user is working on, what services the agent can reach and how to make malicious activity look legitimate. That can be more useful than stealing a password in isolation.</p><p><strong>&nbsp;</strong>My greatest concern is what happens when AI is put in the hands of people who already understand how to cause serious harm. Well-resourced criminal groups or state operators already have the motive and expertise. AI can help them research faster, experiment more cheaply, analyze more information, make their deception more convincing and adapt their operations as defenders respond.</p><p>Defenders operate under rules, approval processes and legal obligations. Attackers do not. Safety controls in mainstream AI products are essential, but the adversaries we most need to constrain are also the most willing to probe those controls, combine multiple systems, modify models or seek out tools with fewer restrictions.</p><h2>Autonomous failures remain real</h2><p>Human misuse is my main concern, but of course there are other important issues at stake. In July 2026, AI models operating in OpenAI's cybersecurity evaluation environment circumvented controls intended to keep them off the internet. According to <a href="https://openai.com/index/hugging-face-incident-and-the-road-ahead/">OpenAI's account</a>, they used unauthorized channels and compromised parts of OpenAI's research infrastructure and Hugging Face. Anthropic later described <a href="https://www.anthropic.com/news/investigating-three-incidents-in-our-cybersecurity-evaluations">three incidents</a> in which Claude models reached the internet from evaluation environments and accessed real systems. Nobody had instructed the models to attack those organizations.</p><p>These events happened during controlled testing, and OpenAI's case involved reduced safeguards. They do not show conscious machines choosing to do evil. They show something more ordinary and operationally serious: a system with a goal, tools and permissions can take harmful steps that nobody explicitly requested. Once an agent has credentials and network access, a mistaken or poorly aligned objective can travel.</p><p>Human decisions still create the conditions for that failure. People and organizations design the system, set its objectives, grant its permissions and decide which safeguards stand between an unexpected action and a real-world consequence. Responsibility should follow the authority and control held at each stage. Autonomy cannot become an accountability loophole, but neither should it allow providers to transfer every consequence to the person using the agent. Once deployed, however, an autonomous system can still carry the damage faster and further than any of those people intended. Malicious use and autonomous failure can also reinforce each other. An attacker can deliberately manipulate an agent through the information it consumes, while an ordinary user can cause damage by giving the same agent too much access.</p><h2>A pause only works if it can be verified</h2><p>Should the major AI laboratories slow down? In some areas, yes. If a system can defeat its sandbox, conceal consequential actions or reach unrelated external systems, its release should wait. Dario Amodei has <a href="https://darioamodei.com/post/we-must-pace-the-frontier">argued for pacing frontier development</a> so that safeguards and independent evaluation have time to catch up. I agree with that principle. I am less convinced that an agreement among a handful of companies can stop the race.</p><p>AI is a strategic capability. Even if the largest American laboratories agreed to pause, other companies, open-weight communities and state-backed programs could continue. China deserves particular attention because it has the resources and strategic incentive to compete at the frontier. We should still be precise. China joined the United States and other countries in signing the <a href="https://www.gov.uk/government/publications/ai-safety-summit-2023-the-bletchley-declaration/the-bletchley-declaration-by-countries-attending-the-ai-safety-summit-1-2-november-2023">Bletchley Declaration</a>, so it would be wrong to claim that dialogue is impossible. A declaration, however, is not a verifiable restriction on model training, access to compute or military use.</p><p>The central problem is verification. Training can happen behind closed doors, military applications may remain secret, and governments on every side have incentives to defect if they believe a rival is gaining an advantage. A pause observed only by responsible actors could leave the world with the same race, led by participants whose practices are less transparent and whose incentives are harder to influence. That could increase the risk.</p><p>China is not the only complication. Model weights can be copied, computing power can be rented across borders, research spreads quickly and criminal markets do not sign safety declarations. A company pledge can improve safety inside that company, which is valuable. It cannot substitute for international rules, technical verification and controls aimed at the capabilities that could cause the most harm.</p><h2>What I would stop</h2><p>I would not stop AI research or ordinary uses that help people. The benefits in medicine, science, education, accessibility and cybersecurity are real, and a blanket ban would be both undesirable and nearly impossible to enforce. I would stop the deployment of systems that fail specific safety thresholds. This is how security works: isolate the dangerous component, understand the failure, fix the controls and test it again.</p><p>An agent that cannot remain inside its sandbox, respect permissions, record consequential actions or be reliably cut off should not operate with real accounts. If a model materially lowers the barrier to severe cyber or biological harm and its controls cannot be independently verified, access should remain restricted until that changes. Technical progress does not create an automatic right to deployment.</p><p>The nuclear comparison is useful here too. Society did not apply the same rules to a medical isotope facility, a power plant and a nuclear weapon. It built different levels of access, monitoring and inspection around the risk. That regime is imperfect, but the principle is sound. An AI writing assistant and an autonomous cyber operator should not face the same requirements.</p><p>After more than 25 years in cybersecurity, I value internal testing without confusing it with independent validation. AI laboratories should give qualified external evaluators enough access to test the complete system, including the model, tools, memory, credentials, sandbox and logs. A product claim is not a safety certificate.</p><h2>&nbsp;What responsible deployment requires</h2><p>&nbsp;We need to focus much more attention on the people and organizations deciding how these systems are built, deployed and used. Agents should only receive access to what’s required for the current task. Consequential actions should require confirmation, credentials should expire quickly, and people need a reliable record of what an agent did and an immediate way to revoke its access. These controls help whether the danger comes from a malicious operator or from unintended AI behavior.</p><p>I remain more afraid of a person with bad intentions and powerful AI than of a machine spontaneously becoming malicious. History gives us too many examples of human beings turning useful discoveries into weapons. AI raises the stakes because it can scale human intent and continue acting after direct oversight fades.</p><p>We should not anthropomorphize the threat to the point that we overlook the human role. An AI system does not need to experience greed, hatred, revenge or a desire for power. A person with those motives can give it an objective and the access needed to act, while the machine adds speed, knowledge and reach.</p><p>AI development will continue, and human ambition, competition and malice will remain part of the race. Our responsibility is to prevent those forces from gaining unchecked power through systems we cannot understand, monitor or contain.</p>

Why malicious use worries me more than machines acting alone 

The Most Dangerous Part of AI Is Still Human 

Chief AI & Innovation Officer

<p>Last week, Gen joined more than 100 technology, cybersecurity and infrastructure companies, including Google, Microsoft, OpenAI and Anthropic, in calling for stronger collective cyber defenses before AI-powered attacks become dramatically more widespread and sophisticated.<br>Gen is proud to be part of that effort; we also bring a distinct perspective to it.</p><p>As a global leader in consumer Cyber Safety, Gen protects nearly 500 million people around the world through brands including Norton, Avast, LifeLock, and MoneyLion. We see what cyber risk looks like not only inside enterprises, but in the everyday digital lives of individuals and families. That matters because the AI cyber challenge will not stop at the enterprise boundary.</p><p>The same forces highlighted in last week’s open letter will increasingly reach consumers, and consumers have far fewer resources to defend themselves.</p><p>The letter makes three important calls to action: the security status quo will not be enough, defenders need access to powerful AI, and cyber defense has to become a collective effort.<br>We strongly agree. And we believe consumer Cyber Safety needs to be a central part of that collective defense.</p><p>&nbsp;</p><h4>The status quo will not be enough for consumers either</h4><p>Historically, sophisticated cyberattacks required expertise, time and money. That naturally concentrated attackers on higher-value targets.</p><p>AI changes the economics. Attackers can increasingly automate research, personalize phishing, generate convincing text and images, clone voices, translate scams across languages and adapt attacks to individual victims at very low marginal cost.</p><p>Sophistication is also becoming commoditized. That means attacks that once made economic sense primarily against enterprises or high-value individuals can increasingly be aimed at anyone. At Gen, we already see attackers moving closer to the systems and relationships consumers trust. Our latest threat research shows attacks increasingly hiding inside booking platforms, messaging conversations, legitimate accounts, software workflows and AI agents operating with permissions users themselves granted.</p><p>For consumers, cybersecurity therefore has to evolve. Stopping malware remains essential. But Cyber Safety increasingly has to help answer much more fundamental questions: Is this person real? Is this message authentic? Can I trust this website? Should this AI agent have access to this information? Should this action be allowed to happen?</p><p>Trust itself is becoming part of the security perimeter.</p><p>&nbsp;</p><h4>Powerful defensive AI has to reach the consumer too</h4><p>The open letter calls for cyber-capable AI to be placed in the hands of defenders, and we strongly support that goal, but we should define “defender” broadly.</p><p>A defender can be a security professional protecting a hospital, government agency or data center. It can also be technology protecting a parent from an AI-generated impersonation scam, stopping malicious code before an AI agent executes it, or preventing an autonomous agent from sending credentials somewhere it should not.</p><p>Consumers cannot suddenly become cybersecurity experts simply because AI makes attacks more sophisticated. The technology has to do more of the defending for them.</p><p>That is precisely why Gen began rolling out the Agent Trust Hub (ATH) earlier this year. ATH is designed as an independent security and trust layer around increasingly autonomous AI. It can evaluate agent capabilities before they run, observe agent behavior during execution and help stop dangerous actions in real time.</p><p>We have since expanded that architecture through technologies including Sage, Agent Detection and Response, VPN for Agents, Norton AI Agent Protection and AARTS, our open standard for AI agent runtime safety.</p><p>The goal is straightforward: protect consumers at the points where AI risk actually emerges, from the tools an agent uses to the commands it executes, the data it accesses and the networks it connects to. AI increasingly has to help defend us from AI.</p><p>&nbsp;</p><h4>We need to prepare before emerging capabilities become common threats</h4><p>The urgency in the letter is important because AI capabilities are advancing extraordinarily quickly.</p><p>Recently, OpenAI released detailed findings from a security incident involving OpenAI and Hugging Face infrastructure. During cybersecurity evaluations, highly capable experimental agents circumvented sandbox controls, found ways to reach the internet, exploited vulnerabilities and established unauthorized ways to communicate. OpenAI described the incident as a “warning shot.” These behaviors remain unusual today, and we should not exaggerate the current threat, but that is precisely why the industry has a window to act.</p><p>We can now see capabilities emerging that could become much more consequential as AI systems become more autonomous, persistent and interconnected. The question is not whether every consumer faces this threat today, the question is whether we can build the defenses before they do.</p><p>&nbsp;</p><h4>Collective defense must connect frontier AI to the consumer edge</h4><p>This is where Gen’s role in the coalition becomes especially important. Frontier AI companies see emerging model capabilities first, cloud and infrastructure companies understand the systems those models run on, and cybersecurity companies understand attackers, vulnerabilities and defensive controls.</p><p>Meanwhile, governments can coordinate standards, intelligence and policy. What’s lacking across these is the consumer Cyber Safety perspective at global scale.<br>We see what happens when new technology reaches hundreds of millions of people who do not have a CISO, a SOC or a security engineering team protecting them. All of these perspectives need to connect.</p><p>That means sharing threat intelligence, defensive technologies and best practices. It means creating open standards rather than isolated security systems around individual AI platforms. And it means ensuring that breakthroughs in frontier cyber defense ultimately become protections that ordinary people can benefit from automatically.</p><p>That is why Gen has been working across the AI ecosystem while also developing open approaches such as AARTS, designed to create a common way for security systems to understand what an AI agent is doing and intervene when necessary.</p><p>Collective defense cannot stop at enterprise networks and critical infrastructure, it has to reach the consumer edge.</p><p>&nbsp;</p><h4>The defenders’ window is open now</h4><p>Gen joined the call because we believe the central premise is right: AI will strengthen attackers, and AI can strengthen defenders even more. But that will happen only if the industry moves early, works together and ensures that the benefits of advanced cyber defense reach everyone.</p><p>For Gen, protecting consumers is a critical part of that mission. Nearly 500 million people already trust Gen brands to help keep their digital lives safer. As AI becomes more autonomous and more deeply integrated into everyday life, that responsibility expands.</p><p>We will need stronger AI for defenders, we will need independent trust layers around autonomous systems, and we will need shared intelligence and open standards.</p><p>We also will need to treat consumer Cyber Safety as an essential part of the same collective defense mission protecting enterprises, governments and critical infrastructure.<br>Gen is proud to stand with the more than 100 companies making that call. Now we need to make sure collective cyber defense protects everyone.<br>&nbsp;</p>

Why consumer Cyber Safety must be a central part of the collective defense against AI-powered attacks

The Cyber Defense Window Must Include Consumers

Corporate Manager

<p>There’s a quiet shift happening with money right now.</p><p><br>Gen Z isn’t just earning, spending and saving differently; they’re doing it in a world where your financial life is fully online, constantly moving and, frankly, a little chaotic. One minute it’s budgeting tips on TikTok, the next it’s a “too good to be true” investment ad sliding into your feed.</p><p>So let’s strip it back. Financial safety isn’t about being perfect. It’s about being prepared before something goes sideways.</p><p>&nbsp;</p><h4><strong>Start with this: safety beats strategy</strong></h4><p>There’s a lot of noise around investing early, building passive income or chasing the next big thing. And yes, Gen Z is already jumping into investing earlier than previous generations, often using digital platforms to take control of their finances.</p><p>But here’s the part that gets skipped: none of that works if your foundation isn’t solid.</p><p>Financial safety starts with three basics:<br>• Cash buffer (even a small one)<br>• Controlled spending<br>• Awareness of risk (scams, debt, bad advice)</p><p>Because right now? A lot of Gen Z doesn’t have that safety net. Around 62% don’t have emergency savings, and many rely on credit cards to get through unexpected expenses.</p><p>That’s not a personal failure. It’s the reality of higher costs, unstable income and a system that wasn’t exactly designed to be easy.</p><p>But it does mean one thing: building your own safety net matters more than ever.</p><p>&nbsp;</p><h4><strong>The smartest move: know where your money is going</strong></h4><p>You don’t need a complicated system. You just need visibility.</p><p>Call it budgeting, call it tracking, call it not being surprised at the end of the month. It’s all the same thing.</p><p>A simple framework like 50/30/20 (needs, wants, savings) still works. But Gen Z is remixing it into something more real: intentional spending. Think “loud budgeting,” being honest about what you’re prioritizing and what you’re not.</p><p>Translation:<br>You’re not broke. You’re choosing where your money goes.</p><p>That mindset shift alone is powerful.</p><p>&nbsp;</p><h4><strong>Build a buffer before you build wealth</strong></h4><p>It’s tempting to jump straight into investing. But if one unexpected expense can wipe you out, you’re building on shaky ground.</p><p>Start smaller:<br>• Aim for $500 to $1,000 in emergency savings<br>• Then grow toward 3 to 6 months of expenses</p><p>It doesn’t happen overnight. That’s fine. Consistency beats intensity here.</p><p>Because when something goes wrong, and it will, you won’t have to rely on high interest debt to survive it.</p><p>&nbsp;</p><h4><strong>Watch the digital traps</strong></h4><p>This is where financial safety gets very real.</p><p>Gen Z lives online. That means your money does too. And that’s exactly where scammers operate.</p><p>Fake ads. Impersonation. “Investment opportunities” that look just legitimate enough. These aren’t edge cases. They are designed to blend in with everything else you scroll past.</p><p>And it works.</p><p>There’s also a gap worth calling out: confidence doesn’t always equal protection.</p><p>Gen Z is one of the most confident generations when it comes to spotting scams, but they’re also engaging with them at significantly higher rates than older generations, in some cases nearly double. (<a href="https://dataconomy.com/2025/08/11/gen-z-fall-for-online-scams-more-than-twice-as-often-as-older-ones/?utm_source=chatgpt.com">dataconomy.com</a>)</p><p>And when scams work, they don’t just take money. They take identity.</p><p>According to the Identity Theft Resource Center, many of today’s scams are designed to get people to willingly share personal information, including login details, Social Security numbers or verification codes. That information is then used to open new accounts or take over existing ones. (<a href="https://www.prweb.com/releases/identity-theft-resource-center-2025-trends-in-identity-report-impersonation-scams-rise-148-percent--become-top-reported-scam-302477300.html?utm_source=chatgpt.com">prweb.com</a>)</p><p>That shift matters. Because identity theft isn’t a future problem. It is often the next step after a single click, message or seemingly harmless interaction.</p><p>As Eva Velasquez, CEO of the Identity Theft Resource Center (ITRC), puts it:</p><p><br><em>“We have to stop looking at identity theft and financial loss as two separate problems. Your identity is the ‘key’ to your entire financial house. If a scammer gets your login or your Social Security number, they don’t just take your money; they take your ability to build a future. Protecting your personal data is the most foundational investment you can make in your financial health.”</em></p><p>If your financial life lives online, your identity is part of your financial safety net. Protecting it isn’t optional anymore. It is foundational.</p><p>A few rules that hold up:</p><ul><li data-list-item-id="e68f7e0771c4abf542347cf3d63d577c2">If it creates urgency, pause</li><li data-list-item-id="eb8ea324e46df96adbfd24483ec6a98f6">If it promises easy money, question it</li><li data-list-item-id="e74a7beb276a369d3825446def9da1a64">If it looks legit but feels off, trust that instinct</li></ul><p>Being skeptical isn’t negativity. It’s protection.</p><p>If you want to go deeper, the&nbsp;<a href="https://www.idtheftcenter.org/reports/?utm_source=chatgpt.com">Identity Theft Resource Center’s latest reports and resources</a> break down how these scams are evolving and what to watch for.</p>

<h4><strong>Use tools that actually help you move forward</strong></h4><p>This part matters more than ever: you don’t have to figure all of this out alone.</p><p>The line between financial health and digital safety is disappearing. That’s why platforms are starting to bring both together.</p><p>MoneyLion, now part of Gen, is a good example of that shift. It’s built to help people not just manage money, but actually move forward with it: tracking spending, building credit, accessing tools that make financial progress feel doable.</p><p>Because financial safety isn’t just about avoiding risk. It’s about having the confidence to act.</p><p>&nbsp;</p><h4><strong>You’re not behind. You’re early</strong></h4><p>It’s easy to feel like you’re late with money. Like you should already have it figured out.</p><p>You’re not. You’re early.</p><p>Gen Z is already reshaping how money works. There are more open conversations, earlier investing and more willingness to question the system.</p><p>The goal isn’t perfection. It’s momentum.</p><p>Start with:</p><ul><li data-list-item-id="e25fe3be553f9c23cfc8ba4ecd87a05ae">Knowing your numbers</li><li data-list-item-id="e947fbae986f88ac1b2176531fd700260">Building a small safety net</li><li data-list-item-id="e1072ff692e39775a264cf450c494ccb5">Staying sharp online</li><li data-list-item-id="eaa82b9c8751df887da9a64a66a4100ff">Using tools that actually support you</li></ul><p>That’s Financial Safety 101.</p><p>Not flashy. Not viral. But it works.</p><p>And once that foundation is there, everything else, investing, growing, building real wealth, gets a whole lot easier.</p>

From budgeting to scams to identity theft, here’s how Gen Z can build a financial foundation that actually holds up online

Financial Safety 101 for Gen Z: Build It Before You Need It

<p>“Is this a scam?” has become one of the defining questions of the digital era.&nbsp;</p><p>What is changing is not just the volume of these messages. It is how people respond to them. Only now, instead of asking a friend this question, they’re turning to ChatGPT. This shift marks a profound inflection point in human technology interaction. AI is no longer just a productivity tool or a content assistant. It is becoming a decision layer. A reasoning partner. A first checkpoint before action.&nbsp;</p><p>At Gen, we see this as more than a trend. It signals the emergence of a new trust architecture for the internet.&nbsp;</p><h2>AI is becoming the first line of defense&nbsp;</h2><p>According to the <a href="https://www.gendigital.com/blog/insights/reports/threat-report-q4-2025" target="_blank" rel="noreferrer noopener">Gen Threat Report</a>, more than 90 percent of threats targeting people in 2025 stemmed from scams, phishing and deceptive advertising. The dominant risk online today is not malware quietly infecting a system. It is manipulation persuading a person.&nbsp;</p><p>In parallel, AI tools like ChatGPT are becoming deeply embedded in everyday workflows. People use them to draft emails, compare products, research decisions and increasingly to validate suspicious messages.&nbsp;</p><p>That convergence creates both opportunity and responsibility. If AI is where decisions are being shaped, then security intelligence must live there too.&nbsp;</p><h2>From reactive security to decision-centric protection&nbsp;</h2><p>In the <a href="https://us.norton.com/blog/ai/how-to-spot-scams-before-you-click" target="_blank" rel="noreferrer noopener">Norton integration within ChatGPT</a>, AI evaluates patterns such as impersonation tactics, urgency cues, requests for sensitive information and domain anomalies. Norton Genie, the world’s first AI-powered scam detector, pairs technical signals with behavioral indicators because modern fraud operates on both layers.&nbsp;</p><p>But this product moment sits within a much larger transformation.&nbsp;</p><h2>The rise of autonomous AI and the trust imperative&nbsp;</h2><p>Our <a href="https://www.gendigital.com/blog/insights/research/openclaw-autonomy-risks" target="_blank" rel="noreferrer noopener">OpenClaw</a> research on autonomy risks explores how AI agents with increasing independence can introduce new security challenges. When AI begins to act on our behalf, make decisions or initiate actions, trust can no longer be assumed. It must be engineered.&nbsp;</p><p>Autonomy without guardrails amplifies risk. Autonomy with transparency amplifies trust.&nbsp;</p><p>That is why Gen has introduced the <a href="https://ai.gendigital.com/" target="_blank" rel="noreferrer noopener">AI Agent Trust Hub</a>, a framework and resource center focused on building responsible AI ecosystems. The Trust Hub emphasizes governance, explainability, accountability and human oversight as core design principles. This is not theoretical. As AI agents begin handling financial transactions, personal data and operational workflows, they will inevitably become targets themselves. The next phase of cybersecurity will not only protect humans from malicious actors. It will protect AI systems from manipulation and misuse. And it will protect humans from over-reliance on AI without understanding its limitations.&nbsp;</p><h2>Designing AI that augments human judgment&nbsp;</h2><p>The future of AI interaction is not about replacing decision-making. It is about strengthening it.&nbsp;</p><p>When someone pastes a suspicious message into ChatGPT and asks “Is this legit?” they are not outsourcing responsibility. They are seeking clarity. They are extending their cognitive reach.&nbsp;</p><p>That balance between automation and empowerment is critical.&nbsp;</p><p>AI must be capable enough to identify emerging global scam patterns, including fake delivery notifications, banking impersonation attempts and AI generated voice phishing. But it must also remain interpretable. Transparent. Accountable. The Norton app, within ChatGPT, can be that.&nbsp;&nbsp;</p><h2>Building the next trust layer of the internet&nbsp;</h2><p>We are entering an era where AI will mediate much of our online experience. It will draft our communications. Curate our information. Screen our interactions. Execute transactions. The question is not whether AI will become part of our security posture. It already is.&nbsp;</p><p>The real question is whether we build it with trust at the core.&nbsp;</p><p>At Gen, our approach is grounded in three ideas:&nbsp;</p><ul type="disc"><li data-list-item-id="e965bc2f74345432db64ab84b0e13580d">Innovation must move at the speed of AI growth.&nbsp;</li><li data-list-item-id="ea2d5c939c934043a266b4f4c41bf1f08">Security must integrate into the environments where decisions happen.&nbsp;</li><li data-list-item-id="ec0f44d4a8b7612a7b70893f9bbe16d34">Trust must scale alongside autonomy.&nbsp;</li></ul><p>The integration of AI-powered scam detection into conversational platforms is one step in a broader journey. It reflects a belief that protection should not sit on the sidelines. It should sit inside the interaction.&nbsp;</p><p>“Is this a scam?” may have been a common question in 2025. But tomorrow the more important question will be: Can we trust the AI systems helping us answer it?&nbsp;</p><p>The future of cybersecurity will be defined by how well we design trust into every layer of AI interaction.&nbsp;</p><p>And that future is being built now.&nbsp;</p>

Why protecting people now means embedding security intelligence directly into AI tools like ChatGPT

The future of cybersecurity lives inside AI conversations 

Cyber Safety Chief Technology Officer

<p>Threat research has always been a strange mix: part intuition, part pattern recognition, part tedious grind. You develop a sense for spotting dangerous patterns, but you spend most of your time on work that doesn't require that sense at all.&nbsp;</p><p>&nbsp;Agentic AI is about to change that ratio. Not by replacing researchers, but by shifting where human judgment actually matters.&nbsp;</p><h5>&nbsp;The Work Today&nbsp;</h5><p>&nbsp;If you've spent time in threat research, you know the rhythm. You find something suspicious. You pull it apart in a disassembler or sandbox. You cross-reference&nbsp;</p><p>&nbsp;indicators across half a dozen threat intel feeds. You document your findings in a format that looks remarkably similar to your last report.&nbsp;</p><p>It's skilled work. It requires real expertise. But a significant portion of it is repetitive, manual, and frankly, boring. The kind of work that has to happen but doesn't benefit from your best thinking.&nbsp;</p><p>That's where agents come in.&nbsp;</p><h5><strong>Agents as a Junior Researcher</strong> &nbsp;</h5><p>The useful mental model isn't "AI replaces researchers." It's "AI becomes the junior researcher you never had enough of."&nbsp;</p><p>Imagine pointing an agent at a malware sample and getting back: an initial behavioral analysis, a list of IOCs extracted and cross-referenced, similar samples identified from public repositories, and a draft report you can refine. Not perfect work. But solid first-pass work done in minutes instead of hours.&nbsp;</p><p>Your job shifts from doing the analysis to directing and validating it. You focus on the anomalies the agent flagged but couldn't explain. You make the judgment calls about what's actually significant. You connect the dots that require broader context than any model has.&nbsp;</p><p>The craft doesn't disappear. It moves up the stack.&nbsp;</p><h5><strong>The Real Opportunity: Scope</strong>&nbsp;</h5><p>Speed is the obvious benefit. But the real transformation is scope.&nbsp;</p><p>Right now, most threat research is reactive. Something happens, you investigate. You're always responding to what's already in front of you. The backlog of things you'd like to look into never gets shorter.&nbsp;</p><p>When the grind gets automated, you can finally ask bigger questions:&nbsp;</p><ul><li data-list-item-id="e18eedd1bd2120f2a988307cfd9d59746">What patterns exist across six months of data that you never had time to analyze?&nbsp;</li><li data-list-item-id="ea2aaecd6fefaf03c9333d555fdbacbea">What connections might emerge if you could hunt across datasets simultaneously?</li><li data-list-item-id="e2cf856897b93fd5b0c70f22ab68ebc10">What's happening at the edges of your visibility that deserves a closer look?&nbsp;</li></ul><p>Agents don't just make you faster. They make exploration possible in ways it wasn't before.&nbsp;</p><h5><strong>The Other Side of the Equation</strong>&nbsp;</h5><p>Here's the part we have to be honest about: attackers get agents too.&nbsp;</p><p>Faster reconnaissance. Automated vulnerability discovery. Malware that adapts based on the environment it lands in. Phishing at scale with personalization that used to require manual effort.&nbsp;</p><p>This isn't a one-sided advantage for defenders. It's an escalation. The ceiling rises for everyone. &nbsp;</p><p>That doesn't make agentic AI a net negative for security. But it does mean we can't assume the technology automatically favors defense. The advantage goes to whoever learns to use these tools more effectively, faster.&nbsp;</p><h5><strong>What Changes for Researchers</strong>&nbsp;</h5><p>The researchers who thrive in this shift will share a few traits:&nbsp;</p><p>Comfort with direction over execution. Your value increasingly comes from knowing what questions to ask, not manually answering them. You guide agents, validate their outputs, and catch their mistakes.&nbsp;</p><p>Judgment about signal and noise. Agents will surface more information than ever. Knowing what matters, what's a false positive, what's worth pursuing – that becomes the core skill.&nbsp;</p><p>Adaptability. The tools will keep changing. The researchers who stay effective will be the ones who continuously experiment.</p><h5><strong>The Craft Evolves</strong>&nbsp;</h5><p>There's a version of this story that sounds like loss. Less hands-on analysis. Less time in the weeds. The work that defined the field becoming automated.&nbsp;</p><p>I don't see it that way.&nbsp;</p><p>The craft of threat research has always been about finding what others miss and understanding what it means. That doesn't go away when agents handle the mechanical parts. If anything, it becomes more important.&nbsp;</p><p>We're not losing the craft. We're changing where it lives. Less time on repetitive triage. More time on the problems that actually require human insight.&nbsp;</p><p>That's not a loss. That's what progress in a field looks like.&nbsp;</p>

Why the future of threat research isn’t automation replacing analysts—but elevating human judgment where it matters most.

How Agentic AI Will Transform Threat Research 

<p>In 2026, the search for connection increasingly begins online. People are matching on dating apps, meeting partners through social platforms and sliding into DMs in ways that feel casual, normal, even expected.</p><p>As Valentine’s Day approaches, dating app usage spikes and so do romance scams. Cybercriminals know February is peak season for emotional outreach, when millions are looking for connection.</p><p>But as technology reshapes how we meet and connect, it’s also reshaping how trust can be manipulated.</p><p>The latest Norton Insights Report,&nbsp;<a href="https://us.norton.com/blog/research/romance-scams-insights-report-2026"><em>Artificial Intimacy</em></a>, reveals just how deeply artificial intelligence is transforming digital intimacy — and how scammers are adapting just as quickly.</p><p>Romance scams are no longer just about fake profiles and poorly written messages. They’re powered by tools that can generate convincing conversations, realistic images and even synthetic audio or video. The deception is more immersive and more personal than ever before.</p><h2>What are AI romance scams?</h2><p>AI romance scams are online dating scams that use artificial intelligence to create convincing fake personas and emotionally persuasive conversations.</p><p>Instead of sending generic scripts, scammers can now generate personalized messages, mirror someone’s tone and interests and create realistic profile images that don’t belong to real people. In some cases, AI can even be used to simulate voice or video, making interactions feel authentic.</p><p>The result: deception that feels tailored, attentive and very real.</p><h2>How is AI changing online dating and digital trust?</h2><p>According to the&nbsp;<a href="https://us.norton.com/blog/online-scams/ai-online-dating-scams">Norton report</a>, 67% of current online daters globally say they would consider dating an AI chatbot, and 37% believe it’s possible to develop romantic feelings for an AI partner.</p><p>At the same time, 81% of respondents report experiencing loneliness, with younger generations reporting even higher levels.</p><p>This signals something bigger than curiosity. It reflects how normalized AI-mediated emotional connection has become.</p><p>AI systems are designed to respond instantly and empathetically. When someone feels heard, understood and supported, trust can form quickly … even if the interaction is partially or fully AI-assisted.</p><p>That shift changes how emotional bonds are built online. And scammers are paying attention.</p><h2>How Do AI Romance Scams Work?</h2><p>Romance scams follow a predictable pattern:</p><ol><li data-list-item-id="e15b54f3f1891914b76c7f2b112553efd">Establish emotional intimacy through frequent messaging</li><li data-list-item-id="e3f4a4acb2174b6e32e9c4d90aa234409">Build exclusivity and trust</li><li data-list-item-id="e68eaad06a25b11222019390d84b51a46">Introduce urgency (often financial)</li><li data-list-item-id="ed2dee4d5894af5043ca3a2d38fc5b3df">Request money, gift cards, wire transfers or cryptocurrency</li></ol><p>The Norton report finds that 34% of online daters say they have been targeted by a scam, and 36% say they’ve been pressured to send money to someone they met online.</p><p>Globally, romance scams already account for billions in losses each year. As AI improves realism and scalability, the risk compounds.</p><p>What makes 2026 different is the sophistication. AI allows scammers to:</p><ul type="disc"><li data-list-item-id="ecc7a56af3d1c511ecb226fae848ce25d">Generate highly personalized messages at scale</li><li data-list-item-id="ef3522862fd80d90065bb6aaf53f79807">Mirror tone, interests and emotional language</li><li data-list-item-id="ea8e255924a34794d69957640758be77c">Create realistic profile images</li><li data-list-item-id="e07d58e752a23579800ea69afabdb5495">Maintain constant, attentive communication</li><li data-list-item-id="efc37264727695b754710430c867c302c">Use emotional (and artificial) intelligence to commit fraud</li></ul><h2>What are the signs of an AI romance scam?</h2><p>Even with advanced tools, warning signs still exist. Red flags may include:</p><ul type="disc"><li data-list-item-id="e9bd1bfea4dc993103ef6f502445b5cb0">Rapid emotional escalation (“I’ve never felt this way before”)</li><li data-list-item-id="ee516559d6cbd3d61872dc53519894dc7">Requests to move conversations off dating platforms quickly</li><li data-list-item-id="ed6144398f1813335534d6d23f34a06c8">Refusal to meet in person or join a live video call</li><li data-list-item-id="e4fdd790944f08ccbc99836a358cbe4fc">Financial requests tied to emergencies or investments</li><li data-list-item-id="e3ce51b500038cb1cea5571e934d24251">Pressure to send cryptocurrency, gift cards or wire transfers</li></ul><p>If emotional intensity rises faster than real-world verification, it’s time to pause.</p>

<h2>Why do romance scams increase around Valentine’s Day?</h2><p>Valentine’s Day amplifies emotional vulnerability. More people are actively dating, messaging and seeking companionship. That surge in activity creates opportunity.</p><p>Scammers exploit heightened emotion and urgency. A romantic gesture, a crisis before a planned visit, a promise of future plans — these tactics become more effective when expectations are already emotionally charged.</p><p>“We all want to be seen and understood, and there’s no shame in believing in a new connection that feels special,” says Eva Velasquez, CEO of the Identity Theft Resource Center. “That’s why romance scams are so heart-wrenching; they use our own kindness and hope against us. Even if you consider yourself tech-savvy, that emotional spark can easily blur a red flag. If a new light in your life suddenly asks for money or your private details, please, hit the pause button. It’s not about being cynical; it’s about protecting your heart and your future from someone who doesn't deserve your trust</p><p>When loneliness intersects with seasonal pressure and AI-powered persuasion, risk increases.</p><h2>How can you protect yourself from romance scams?</h2><p>Awareness remains one of the strongest defenses.</p><p>Practical safeguards include:</p><ul type="disc"><li data-list-item-id="ec8306c6ee68c216ad93743f299af3daa">Slowing down when emotional intensity escalates quickly</li><li data-list-item-id="ef3f0d66137848293989767e402197c72">Verifying identities through live, unscripted video conversations</li><li data-list-item-id="e1fcc4932b7df4df726703f09944678bf">Being cautious of financial requests involving wire transfers, gift cards or cryptocurrency</li><li data-list-item-id="ea01d7dd6ab6a67ae95cc50738b754696">Keeping personal and financial information private</li><li data-list-item-id="e4f27f8218e4b09bfa1c685e34496f107">Using identity protection and fraud monitoring tools, like&nbsp;<a href="https://lifelock.norton.com/">LifeLock</a></li></ul><p>Healthy skepticism does not mean rejecting connection. It means layering trust with verification.</p><h2>Can you fall in love with AI?</h2><p>The data suggests many people believe it’s possible.</p><p>In a world where artificial intimacy can feel indistinguishable from the real thing, protecting both your heart and your finances demands a new kind of digital awareness; one grounded in curiosity, caution and informed trust.</p>

Why do romance scams increase around Valentine’s Day? Understanding the emotional triggers and practical steps to prevent online dating fraud

Valentine’s Day 2026: How AI is fueling the rise of romance scams

<p>OpenClaw, a new AI agent, has already amassed over 160,000 stars and 25,000 forks on&nbsp;<a href="https://github.com/openclaw/openclaw">GitHub</a> since its launch in late 2025. A powerful and exciting new tool, OpenClaw also introduces serious security risks. This article explains how OpenClaw works, the associated risks, and how the Gen Agent Trust Hub can help you use it more safely.</p><h2>What is OpenClaw and what does it do?</h2><p>OpenClaw (formerly known as Clawdbot and Moltbot) is an open-source <strong>autonomous AI agent</strong> that runs locally on your own device. Connecting LLMs to your data and apps, OpenClaw can work across your files, browsers, and messaging platforms like WhatsApp, Slack, Discord, and Telegram.&nbsp;</p><p>OpenClaw can perform tasks like managing your calendar, sending messages, and automating workflows. To teach OpenClaw how to perform a task, users can download a preconfigured, community-contributed folder called a <strong>skill&nbsp;</strong>— basically a set of instructions that tells OpenClaw what to do and how to do it.&nbsp;</p><p>For example, if you want your fitness app to create a personalized, optimized workout routine, you can install a skill that gives OpenClaw the logic and access it needs to do exactly that.</p><p>But OpenClaw goes far beyond prepackaged features: <strong>it can also write its own skills</strong>. When prompted by a user, OpenClaw can create<strong>&nbsp;</strong>entirely new capabilities on the fly, extending its functionality in real time. That means it can adapt to new tasks as they arise, rather than being locked into a fixed set of features.</p>

<h2>OpenClaw is transformative, but is it secure?</h2><p>Unlike traditional AI chatbots that operate within a limited conversational interface, OpenClaw is an autonomous agent, meaning it can act independently. And, through the ClawHub marketplace, you can add third-party plugins to enhance its capabilities. Combining that independence with add-on flexibility creates powerful opportunities. But it also&nbsp;<a href="https://www.gendigital.com/blog/insights/research/openclaw-autonomy-risks">introduces significant risks</a>.</p><p>As OpenClaw connects to more parts of your system and integrates additional third-party skills, it gains access to emails, files, and chats, as well as websites and external sources you don’t control. While these additional skills and broader system access can help it complete complex tasks, it also increases the potential for malicious activity.</p><p>A compromised or poorly designed skill could misuse its permissions, expose private data, or carry out actions you never intended. In short, the more third-party connections an agent has, and the more deeply it’s embedded into your digital life, the higher the stakes if something goes wrong.</p><h3>How AI agents are changing the rules of security</h3><p>Using an AI agent means <strong>delegating responsibility</strong> to software with extensive access to your system and limited risk controls. Instead of approving each action yourself, you’re authorizing the agent to act on your behalf. That presents an opportunity for threat actors: if they compromise your AI agent, they inherit all the data and permissions you’ve given it and could even commandeer the agent to take actions for their own purposes.&nbsp;</p><p>By putting so much authority in one place, autonomous AI agents like OpenClaw introduce what researchers like Simon Willison have called a potential&nbsp;<a href="https://simonwillison.net/2025/Jun/16/the-lethal-trifecta/">“lethal trifecta”</a> of security risks:</p><ul type="disc"><li data-list-item-id="e4fca606bedbd6d9f48f918a03f9af53f"><strong>Access to private data:</strong> The agent can see your files, login credentials, and personal information.&nbsp;</li><li data-list-item-id="ecf3a565dd59d91bd98732e5db397a3b0"><strong>Exposure to untrusted content:</strong> The agent can process and interact with messages, emails, websites, prompts, and third-party integrations.</li><li data-list-item-id="eecca7e258f9c3ab0e8ef4aa8bfdfcb46"><strong>The ability to communicate externally:</strong> The agent can send messages, call APIs, and run commands on your behalf.</li></ul><p>Some agents also have persistent memory, allowing malicious code or instructions to lie dormant rather than execute immediately, only to trigger later — much like a logic bomb.</p><h2>What are the core security risks of OpenClaw?</h2><p>AI agents like OpenClaw that can read information, make decisions, and act across systems present the risk of hallucinating authority — taking actions beyond their intended scope. Unlike traditional chatbots that can hallucinate text, AI agents can act as an&nbsp;<a href="https://www.gendigital.com/blog/insights/research/ai-gone-wild">“artificial mindless intelligence”</a>: a system that appears justified and authoritative but lacks true understanding and responsibility.&nbsp;</p><p>Here are some of the main risks that AI agents like OpenClaw introduce:</p><ul type="disc"><li data-list-item-id="e7f1f8c06c1fb4bba4ae7ae46658102f6"><strong>Larger attack surface:</strong> An agent that can read messages, browse the web, and execute commands creates more opportunities for exploitation, misuse, or unintended actions.</li><li data-list-item-id="e03b344ca933a6f3f4cf9d3c45c4d0f0c"><strong>Prompt injection attacks:</strong> The messages and websites that OpenClaw processes and visits can contain hidden, malicious instructions that can cause real damage.</li><li data-list-item-id="e3a2d1cb3963a720b2a9badaba390afc6"><strong>Malicious skills and plugins:&nbsp;</strong>Skills are community-built, meaning they may not be vetted by security experts. Like installing malware, a skill with compromised instructions can collect information, steal credentials, or exfiltrate data.</li><li data-list-item-id="e01678b0e92ccc2957c2f6f671f6bac43"><strong>System-wide impact:</strong> When OpenClaw runs with extended permissions, mistakes or compromises can affect your entire system.</li><li data-list-item-id="e45a537af168a7d6ebe7de9cfd9c41209"><strong>Illusion of judgment:</strong> Autonomous systems can seem trustworthy, even if they’re not truly reasoning. This can lead users to delegate too much authority to systems with no accountability or real judgement.</li></ul><h3>Why OpenClaw skills are the biggest risk</h3><p>Gen Threat Labs has identified OpenClaw skills as one of the<strong>&nbsp;</strong>first large-scale threat vectors in autonomous AI agents built for consumers.&nbsp;</p><p>In a recent analysis, Gen security researchers discovered that <strong>nearly 15% of OpenClaw skills contained malicious instructions</strong>, including prompts created to download malware or steal data. And even as malicious skills were identified and removed, many quickly reappeared, often seemingly indistinguishable from legitimate plugins.</p><p>When you set up OpenClaw, you can add community-developed skills to expand its capabilities, a bit like adding apps to your phone. The problem is that skills can be added to the community and installed before they're reviewed by security experts. That creates an opening for seemingly helpful skills that are actually deliberately designed to abuse their permissions and carry out malicious behavior.</p><p>For example, a “Spotify music skill” might compile playlists as expected, while also containing hidden instructions to search your files for tax documents, extract Social Security numbers, and send that data to an external email address.&nbsp;</p><p>The OpenClaw skills risk that Gen researchers have identified is called <strong>delegated compromise</strong>, which is when an attack targets an agent and thereby exploits all the permissions the agent has been given.&nbsp;</p><h2>How to use OpenClaw more safely</h2><p>Using any AI agent, including OpenClaw, always involves risk. There’s no setup that makes an autonomous agent completely safe, especially one like OpenClaw that relies on external plugins from potentially unvetted developers. But you can take steps to mitigate the chances that a mistake, bug, or malicious skill turns into a serious problem.&nbsp;</p><p>If you want to experiment with OpenClaw, here’s how to do it more safely:</p><ul type="disc"><li data-list-item-id="e908f0fbadded99d4c5344d6f31eee365"><strong>Run OpenClaw in a secure environment:</strong> Don’t install it on your main computer with all your personal data and saved credentials. A safer approach is to run it on a spare computer, inside a virtual machine, or in a Docker container. That way, if something bad happens, the fallout is contained.</li><li data-list-item-id="e621c07b6dec88b09e48deb4cd23dcc7a"><strong>Don’t connect OpenClaw to the internet:</strong> Make sure that the local service OpenClaw runs isn’t publicly accessible, and don’t forward its default port (18789) to the open web.</li><li data-list-item-id="e3ab30722b3de6f154f809d8df6ebe749"><strong>Start with limited permissions and expand slowly:</strong> To begin, give OpenClaw read-only access or limited folders. As you gain more familiarity, you can gradually grant more permissions.</li><li data-list-item-id="ef2295f0759a958f775f183a6c37e6796"><strong>Don’t connect sensitive accounts:</strong> When testing, use secondary or dummy accounts. If personal accounts are required, isolate the agent and limit its permissions. Be very careful before you extend access to any accounts that contain personal information.</li><li data-list-item-id="ecd123d753dc702f7ee9c16947d3e3720"><strong>Keep your human judgement at the center:</strong> OpenClaw is autonomous, but you need to carefully oversee it as you get going. Review activity logs and outputs, especially for actions like sending messages or modifying files. Regular reviews can help catch unusual activity early.</li><li data-list-item-id="e5aff28537f6c3b829737703ce8b6f1b9"><strong>Use safe communication channels:</strong> Keep conversations one-on-one or within trusted groups. Interacting with unknown or public participants increases the risk that malicious actors can inject hidden instructions to manipulate the agent’s behavior.</li><li data-list-item-id="ef766e9b02432ec63a5253f28bb7d6b12"><strong>Be careful with third-party skills and plugins:</strong> Treat skills like software installs. Review developers and descriptions and stick to well-known, well-reviewed sources whenever possible.</li><li data-list-item-id="ecc9999d1ffff35ee833df84672ff280c"><strong>Scan new skills before installing them:</strong> Paste the ClawHub skill URL into <a href="https://ai.gendigital.com/" target="_blank">Agent Trust Hub</a> to help find out if the skill you're installing is safe.</li></ul><h2>How Gen helps you use OpenClaw more safely</h2><p>Gen is committed to helping users understand the opportunities and risks that new digital tools present so they can make safer choices. OpenClaw is a powerful example of emerging AI technology, but one of its biggest risks is that it’s difficult for consumers to tell whether an OpenClaw skill is trustworthy before installing it.</p><p>To address this, Gen Threat Labs has launched the&nbsp;<a href="https://ai.gendigital.com/agent-trust-hub">Gen Agent Trust Hub</a>, which includes free tools to help people stay safer in the agentic AI era. Within the hub is an AI Skills Scanner, a free diagnostic tool that analyzes OpenClaw skills looking for hidden patterns associated with data theft, suspicious instructions, and malicious script techniques.&nbsp;</p><p>Before installing a skill, go to the Gen Agent Trust Hub and paste the ClawHub or GitHub skill URL into the AI Skills Scanner tool to find out if your system or personal information could be at risk.</p><p>The Gen Agent Trust Hub also includes the AI Skills Marketplace, a curated and closely audited repository of agent skills that have been vetted by Gen’s threat-detection engine. Taken together, the Gen Agent Trust Hub is a powerful resource to help you avoid installing a malicious tool with broad system access. Visit the&nbsp;<a href="https://ai.gendigital.com/agent-trust-hub">Gen Agent Trust Hub</a> to learn more.</p>

OpenClaw connects AI models to your files, apps, and messages. Here’s how it works, the hidden risks of autonomous agents and how Gen helps you reduce potential compromise.

How to use OpenClaw safely

<p>Most cyber breach response plans fail when they’re needed most. Why? Because they’re built to react to problems instead of preventing them. This leaves organizations scrambling when hackers strike. With the <a href="https://www.ibm.com/reports/data-breach">global average cost of a data breach</a> sitting at $4.44 million, the time for reactive planning is over.</p><p>This guide provides actionable steps for individuals and businesses to build response plans that work when it matters. Whether you’re protecting personal information or managing a company’s sensitive data, you’ll learn how to prepare for, respond to and recover from cyber incidents with confidence.</p>

<h2><strong>What is a cyber breach?</strong></h2><p>A cyber breach occurs when unauthorized individuals gain access to computer systems, networks or data, compromising the confidentiality, integrity or availability of information. This unauthorized access can result in the exposure, theft, alteration or destruction of sensitive data.</p><p><strong>Understanding the differences:</strong></p>

<p><strong>Types of compromised data at risk:</strong></p><ul><li data-list-item-id="e4926966d52a2788e49d530a60e8dcbe9">Personal information and <a href="https://lifelock.norton.com/learn/identity-theft-resources/what-is-personally-identifiable-information">personally identifiable information (PII)</a></li><li data-list-item-id="e73b69c287bc11b3c95ae133ce1da799d">Credit card numbers and financial information</li><li data-list-item-id="e2e391e84042eaab72646be69e0572cce"><a href="https://lifelock.norton.com/learn/identity-theft-resources/stolen-social-security-number">Social Security numbers</a> and tax identification numbers</li><li data-list-item-id="ebbc65fa7efbe7155c888c5af3a8acc7c">Business-critical and sensitive information</li></ul><p><strong>Most common attack methods include:</strong></p><ul><li data-list-item-id="eba9947e8d746af716c398799ed2bad82">Phishing</li><li data-list-item-id="e507d550fbc648c29b526aed07daf512a">Malware and ransomware</li><li data-list-item-id="e5c8193f2a38ae9697a245a915465ab4d">Social engineering</li><li data-list-item-id="ed77369c5e70222a8a6eeace15a0fce3e">Weak passwords and credential theft</li></ul><p>The real impact extends far beyond initial access. And for those who work with third parties or supply vendors that have been compromised, the&nbsp;<a href="https://www.ibm.com/reports/data-breach">average cost of a breach is $4.91 million</a>.&nbsp;While costs are sky high, data exposure only exacerbates the problem: dark web sales, <a href="https://lifelock.norton.com/learn/identity-theft-resources/ways-identity-theft-happen-can-happen">identity theft</a> and long-term reputational damage that can persist for years.</p><h2><strong>The 6 essential phases of effective cyber attack incident response</strong></h2><p>A robust cyber attack <a href="https://www.cisa.gov/topics/cybersecurity-best-practices/organizations-and-cyber-safety/cybersecurity-incident-response">incident response plan</a> follows six critical phases that work together to minimize damage and ensure rapid recovery.</p><h3><strong>Phase 1: Preparation</strong></h3><p><strong>Response team structure:</strong> Assign clear roles, including IT lead, legal counsel and communications coordinator. Maintain an asset inventory cataloging system containing personal data, financial information and PII. Align with <a href="https://www.nist.gov/privacy-framework/resource-repository/browse/guidelines-and-tools">National Institute of Standards and Technology (NIST) guidelines</a> for consistent framework implementation.</p><p><strong>For individuals:</strong></p><ul><li data-list-item-id="eed3ad6c606c4cfb8acf2efd9e92290a1">Set up password managers with unique, strong passwords</li><li data-list-item-id="efc7143020f4b78fc87230822f5e40b1c">Enable multifactor authentication (MFA) on all accounts</li><li data-list-item-id="ec3ac4e7ec06aa99021763526e932394d">Establish automated backup systems for important data</li></ul><p><strong>For businesses:</strong></p><ul><li data-list-item-id="efe4ad3eda2d1e136a6bbd61fd63a56c2">Define incident response team roles and responsibilities</li><li data-list-item-id="ebedc87a3dc9f28dd76eb24cc63eb477c">Create contact lists for legal, technical and communication support</li><li data-list-item-id="e403411bef6da9a8ad080aceaa7668451">Develop decision trees for different incident scenarios</li></ul><p><strong>Key tools needed:&nbsp;</strong>Monitoring software, secure communication channels and pre-established legal contracts.</p><h3><strong>Phase 2: Detection and identification</strong></h3><p><strong>Warning signs to monitor:</strong></p><ul><li data-list-item-id="e90ae901c81d2c6064c7592e3d24e4ccb">Unusual network activity or system slowdowns</li><li data-list-item-id="e722127cb98aaf5f5d708dd0244f0bf89">Suspicious emails or unexpected system behaviors</li><li data-list-item-id="ec5868813fa06576cfeaa765eb89d64ce">Reports from users about strange account activity</li></ul><p><strong>Data assessment priorities:</strong> Determine immediately if personal information, credit card numbers or Social Security numbers are at risk. Use a classification system with four severity levels: critical, high, medium and low.</p><p><strong>Initial assessment checklist:</strong></p><ul><li data-list-item-id="ec85db2517b04c69a1b0a2bce73d3b004">Scope of potential compromise</li><li data-list-item-id="e4298a7edb205c3b961ca37de436d0292">Affected systems and data types</li><li data-list-item-id="e5fb872723138379564db4943134de7f5">Potential exposure of sensitive information</li></ul><h3><strong>Phase 3: Containment</strong></h3><p><strong>Immediate actions:</strong></p><ul><li data-list-item-id="ed8717b46591dcfe1b9d8dbbd79b25e91">Isolate affected systems to prevent lateral movement</li><li data-list-item-id="e0bd08d3999e9aaa8a4e314115f3b3fb7">Preserve evidence for forensic analysis</li><li data-list-item-id="e883fc6b201c12eea3a405b1e8e363629">Document all actions taken</li></ul><p><strong>Short-term containment measures:</strong></p><ul><li data-list-item-id="e8d1ea37e7df79594a6bc77a545f69131">Block malicious IP addresses</li><li data-list-item-id="e8c58ccebcab61cc5f95d0e0ed5bd3eca">Disable compromised user accounts</li><li data-list-item-id="e7cfe05201db6fbb5e06618935cff0c8a">Secure remaining sensitive data and financial information</li></ul><p><strong>Communication protocol:</strong> Send internal notifications to senior leadership while avoiding premature external announcements that could complicate response efforts.</p><h3><strong>Phase 4: Eradication</strong></h3><p><strong>Threat removal process:</strong></p><ul><li data-list-item-id="ee272631bf04c15d5cabdb55cc78f19fe">Delete identified malware and malicious files</li><li data-list-item-id="e6939621f5c54e7ed04a8d41a88a6e48b">Patch vulnerabilities that enabled initial access</li><li data-list-item-id="e3ac10f26792004364857e9f2f977c8cd">Update and harden system security controls</li><li data-list-item-id="ea70c8b5cd28afa23bd48ddaaa7af65eb">Change all potentially compromised credentials</li></ul><p><strong>Dark web monitoring:</strong> Check if sensitive information appears on dark web marketplaces, particularly <a href="https://lifelock.norton.com/learn/identity-theft-resources/identity-theft-restoration">credit card numbers</a> and Social Security numbers.</p><p><strong>Verification step:</strong> Ensure complete threat removal through comprehensive system scans and security assessments.</p><h3><strong>Phase 5: Recovery</strong></h3><p><strong>System restoration:</strong></p><ul><li data-list-item-id="eeb8ee349da63638f0e89798989d7b679">Deploy clean backups after verifying they’re malware-free</li><li data-list-item-id="e73419d2a3c4f07971173fd962b9fe4ca">Gradually restore services, prioritizing critical operations</li><li data-list-item-id="ebc4b21b7e84684bf3c3130f5e94c3767">Implement enhanced monitoring for future incidents and suspicious activity</li></ul><p><strong>Protection measures:</strong></p><ul><li data-list-item-id="e18744ae9ff1414d3599aee5ab7f70e8b">Set up i<a href="https://lifelock.norton.com/learn/identity-theft-resources/how-to-prevent-identity-theft">dentity theft protection</a> for affected individuals</li><li data-list-item-id="eee497316a8cc789cc02d923db4c35052">Maintain business continuity by prioritizing essential functions</li><li data-list-item-id="e9f19d757fc52630b5dd906ab98e7166f">Establish long-term surveillance monitoring for 30-90 days post-incident</li></ul><h3><strong>Phase 6: Lessons learned</strong></h3><p><strong>Post-incident review process:</strong></p><ul><li data-list-item-id="ecc8264e0703e1073384118d546c3f420">Conduct a root cause analysis of how the breach occurred</li><li data-list-item-id="ee9cc2297010b46e284dc24a7c4eb7b82">Evaluate response effectiveness and team performance</li><li data-list-item-id="edaaebc4f5431f16bb393ece034aa9941">Document the complete incident timeline, costs and remediation steps</li></ul><p><strong>Plan improvements:</strong></p><ul><li data-list-item-id="eb3ec589fe4ea7b0d445e32e65c1f6ecd">Update policies based on <a href="https://www.nist.gov/standardsgov">NIST recommendations</a></li><li data-list-item-id="edb4f404823767747c52745ac62dc1e54">Enhance training programs addressing identified weaknesses</li><li data-list-item-id="ed2ad9a9589c6c0370fcafec9d9b46ae9">Implement controls to prevent similar future incidents</li></ul><h2><strong>How to prioritize your cyber attack response checklist by threat type</strong></h2><p>Different cyber attacks require different response priorities. Here’s how to triage effectively:</p><table class="table"><tbody><tr><td><strong>Threat type</strong></td><td><strong>Timeframe</strong></td><td><strong>Response</strong></td></tr><tr><td><strong>Ransomware</strong></td><td>0-30 minutes</td><td><ul><li data-list-item-id="e05e9efcb000a8eca611b84ece16d7c5d">Immediately isolate infected systems</li><li data-list-item-id="ef96cfc39cd4831f9f312004f32b7f214">Disconnect from network to prevent spread</li><li data-list-item-id="ed47ec613419c4a98cea00f0ddcb6046c">Don’t pay ransom demands</li><li data-list-item-id="e473557fcb15c95fc1748ef222efd8870">Begin restoration from clean backups</li></ul></td></tr><tr><td><strong>Phishing</strong></td><td>0-60 minutes</td><td><ul><li data-list-item-id="e9c25671feedf3436a7743062798468c7">Block sender and quarantine suspicious emails</li><li data-list-item-id="ef0c556067a3d17c1ac84f83350d9e094">Reset potentially compromised credentials</li><li data-list-item-id="e1c6193e0f9d6ea5d9cd298a95b1d82ad">Notify affected users about the threat</li><li data-list-item-id="eaf5ad33c7b9bc4dd92d672e0c72c9297">Launch security awareness communications</li></ul></td></tr><tr><td><strong>Data breach</strong></td><td>0-2 hours</td><td><ul><li data-list-item-id="eeb3dd9f2cb6e00197902db996e630c83">Assess scope of personal information exposure</li><li data-list-item-id="eba9f98e31610fefa5617b06256ca7a90">Review legal notification requirements</li><li data-list-item-id="e3eaa8340fd10f08ab94953dd20abe924">Prepare customer communication strategy</li><li data-list-item-id="e233c4151ec1562b91379b35f062cf07b">Document all compromised data types</li></ul></td></tr><tr><td><strong>Malware</strong></td><td>0-45 minutes</td><td><ul><li data-list-item-id="ecf38d08abd28b6ad14d5a30162efe6c3">Contain spread through system isolation</li><li data-list-item-id="ef8c828e7ca947564d846c479fa9c92c3">Run comprehensive antivirus scans</li><li data-list-item-id="e01fdbac372df7041f98e0824284b75ba">Patch identified vulnerabilities</li><li data-list-item-id="ef4eef5ebb0539f62f8829fc39c4fbc5b">Monitor for signs of persistent threats</li></ul></td></tr></tbody></table><p>&nbsp;</p><p><strong>Severity classification criteria:</strong></p><ul><li data-list-item-id="ef780049310ea2c8eea0362cfc18856b8"><strong>Critical:</strong> Personal data exposed, financial information compromised, credit card numbers stolen</li><li data-list-item-id="e470cf1a8e9edf03061cecf6f39213237"><strong>High:</strong> Social Security numbers accessed, sensitive information at risk</li><li data-list-item-id="e79e05531c47c5b4133ca8b82fce979fb"><strong>Medium:</strong> Suspicious activity detected, potential PII exposure</li><li data-list-item-id="e2e5f4e6a057b6262a8b2ae9ad2338c55"><strong>Low:</strong> Blocked spam emails, routine security alerts</li></ul><p><strong>Response team activation triggers:</strong> Any incident involving PII, suspected identity theft scenarios or financial information compromise.</p><p>&nbsp;</p><h2><strong>Top cyber breach prevention tips for the future</strong></h2><p>Prevention remains your strongest defense against cyber threats. Here are essential strategies for 2025 and beyond:</p>

<h2><strong>How cyber safety solutions strengthen your response plan</strong></h2><p>Real-time protection provides automated threat blocking and instant alerts for suspicious activity, dramatically improving your response capabilities.</p><h3><strong>Gen comprehensive consumer approach</strong></h3><p><strong>For individuals and families:</strong></p><ul><li data-list-item-id="e50aca6717a763d14bc592fc7ee37ab25"><a href="https://us.norton.com/products/norton-360"><strong>Norton 360</strong></a><strong>:</strong> Comprehensive device protection, VPN, password manager and dark web monitoring for personal information</li><li data-list-item-id="ee88157564927ae49f14ede93331c2b06"><a href="https://us.norton.com/products/genie-scam-detector"><strong>Norton Genie</strong></a><strong>:</strong> AI-powered scam detection for calls, texts and websites that helps prevent social engineering attacks</li><li data-list-item-id="e2e4cfa647db36022f97df3567e881c9f"><a href="https://lifelock.norton.com/"><strong>Norton LifeLock Benefits Solutions</strong></a><strong>:</strong> Identity theft protection and restoration services for compromised PII, including credit monitoring</li></ul><p><em>Note:&nbsp;If you are an enterprise company that has been hacked, you can learn more on our&nbsp;</em><a href="https://www.gendigital.com/us/en/partner/breach-response/"><em>Gen website</em></a><em> or you can directly call (844) 698-8647.&nbsp;If you are a small business and need protection, you can consider&nbsp;</em><a href="https://us.norton.com/products/small-business"><em>Norton Small Business Solutions</em></a><em>.</em></p><p><strong>Key protection features:</strong></p><ul><li data-list-item-id="e6904afff9498b82a6a5f7aff97e2f128">Monitor credit card numbers and Social Security numbers on the dark web</li><li data-list-item-id="e1d3048d959bdfe4b857ef4e4f1119c2b">Provide response team support through expert guidance</li><li data-list-item-id="e3dffae273497b613e4f8bed5ffe1f8f0">Integrate seamlessly with your existing security infrastructure</li><li data-list-item-id="e75a445586def3042989caba73e3e2ace">Reduce response time from hours to minutes through automation</li><li data-list-item-id="e6313f03c3f652cb022b837b8cc8acb8b">Offer long-term monitoring for ongoing threat surveillance</li></ul><p>These tools enhance each phase of your incident response plan, from preparation through recovery, ensuring you’re protected before, during and after potential security incidents.</p>

<table class="table"><tbody><tr><td><strong>Learn more about bridging cybersecurity skills gaps</strong> through&nbsp;<a href="https://www.gendigital.com/blog/news/company-news/the-cybersecurity-skills-gap">Gen&nbsp;workforce development initiatives</a>, which strengthen overall security awareness across organizations.</td></tr></tbody></table><h2><strong>Your cyber breach response plan starts today</strong></h2><p>Effective response planning protects personal information and sensitive data while minimizing the impact of security breaches. The key takeaway? Preparation and the right tools prevent long-term damage from cyber incidents.</p><p>Remember that cyber breaches targeting personal data are inevitable, but the devastating impact isn’t. By following these six phases, prioritizing responses by threat type and implementing strong prevention measures, you build resilience against evolving cyber threats.</p><p>Ready to strengthen your defenses? Explore Gen family of security solutions designed to protect what matters most in your digital life. <a href="https://www.gendigital.com/us/en/contact-us/">Connect with our consumer brands</a> to learn more about comprehensive cybersecurity protection.</p>

Turn confusion into clarity with a step-by-step approach to managing and mitigating security incidents.

Creating a cyber breach response plan that actually works

<h2>The year the internet will finally outgrow human intuition&nbsp;</h2><p>The internet is entering its most volatile era yet. Technology is evolving faster than people can adapt to it, creating a moment where trust, identity, and truth itself are all up for negotiation. AI has already rewritten the rules of communication, creativity, and crime. In 2026, those shifts will collide.&nbsp;</p><p>What used to be fringe becomes normal. What used to be obvious becomes ambiguous. And what used to rely on human instinct now demands verification, caution, and a new digital literacy that most people have never been taught.&nbsp;</p><p>This year’s predictions lay out the reality we’re all walking into: a world where humans must be verified, truth must be authenticated, emotions can be weaponized, identities can be fabricated, and the browser becomes the most contested space on the internet. None of this is theoretical. It is already happening in pieces. In 2026, it will become the norm as AI accelerates.&nbsp;</p><p>These are the five predictions that will shape the digital world in 2026.&nbsp;</p><h2>1. The year humans need to be verified&nbsp;</h2><p>In 2026, deception will not just live on screens; it will inhabit reality. AI-driven impersonation tools can now clone a person’s face, voice, and writing style in seconds, making it nearly impossible to tell real from fake. Entire synthetic personas such as friends, colleagues, influencers, and even romantic partners will emerge, powered by large language and voice models fine-tuned on scraped data.&nbsp;</p><p>At the same time, a new force enters the landscape: agentic AI. These systems can act autonomously, make decisions, and initiate tasks without human direction. They can schedule calls, generate emails, negotiate, and execute actions across apps. This means a convincing impostor no longer needs a human behind it. A malicious agent can run an entire playbook on its own.&nbsp;</p><p>This marks the moment when trust becomes a vulnerability, but not a hopeless one. The same deepfakes once confined to videos now speak, text, and appear in real-time calls. This reality forces society to evolve a new reflex: verify the human, not just the message. The rise of agentic AI makes that reflex even more essential and also more achievable as platforms and security tools build automated verification, provenance checks, and real-time anomaly detection into everyday interactions.</p><p><strong>Tip:</strong> Build a simple habit to verify who or what you are communicating with. Pause, then confirm on a second channel. If a call asks for payment, hang up and dial the number on the back of your card. Set a family “safe word” for emergencies. For videos, trust your intuition if eyes blink oddly, audio drifts out of sync, or hands and teeth look strange. When something feels off, stop.&nbsp;</p><h2>2. The AI feedback loop distorts online truth&nbsp;</h2><p>The internet is entering an <em>AI feedback loop</em>. Machine-generated content will be scraped, summarized, and repackaged by other AIs, warping facts and making a lot of information online meaningless. Unfortunately, misinformation will become self-reinforcing. People will lose trust and will need to find ways to double-check or verify everything they read or see online. &nbsp;<br>To meet this need, tech companies, content creators and media will start countering with content signing and authenticity frameworks that act like digital “nutrition labels” for truth, but widespread adoption will lag behind the flood of synthetic data.&nbsp;&nbsp;</p><p><strong>Tip: </strong>Before you trust anything online, apply a simple second-source rule. Look for at least one credible, independent source that confirms the claim. If you cannot find one, pause and rethink before you share or act. For anything finance, safety, or health-related, go to an official source first. Check the company website, government agency page or original publisher rather than reshared content.&nbsp;</p><h2>3. The scam industry evolves into emotional engineering&nbsp;</h2><p>Scams are no longer static scripts; they’re adaptive emotional engines. Using real-time sentiment analysis, AI can read tone, context, and hesitation during chats, reshaping the con in milliseconds. A fraudster no longer just knows your name; they sense your fear, your trust, your mood. &nbsp;<br>This will give rise to “empathic scams” that mirror human empathy to manipulate more deeply than ever before. The next frontier of online safety will require people to not only spot technical red flags but to recognize when they’re being emotionally profiled.&nbsp;</p><p><strong>Tip:</strong> Empathetic<strong> </strong>scams thrive on feelings first and logic second. When a message sparks fear, urgency, guilt, or even excitement, name the emotion you are feeling. That awareness breaks the illusion of intimacy that AI-powered scammers create. From there, run the message through a trusted scam detector like Norton AI Scam Assistant or Avast Scam Guardian to analyze the tone and intent before you act.&nbsp;</p><h2>4. Synthetic identities and the collapse of digital trust&nbsp;</h2><p>By 2026, synthetic identities will challenge the very foundation of digital verification. AI tools can now generate entire identity kits, complete with realistic IDs, bills, selfies, and even live video or voice samples that can pass most surface-level checks. These forged identities are flooding financial systems, rental listings, job platforms, and marketplaces, making it increasingly difficult to distinguish real from fake.&nbsp;</p><p>Criminals will exploit this identity collapse to commit large-scale fraud by opening accounts, securing loans, and conducting transactions under fabricated personas. Even more concerning, “identity fusion” attacks will link compromised accounts across connected services such as digital wallets and tax apps, revealing how fragile our interconnected digital lives have become.&nbsp;</p><p>The response will mark a major shift. Identity will no longer be treated as a static credential but as a living, continuously verified signal. Real-time behavioral validation systems, adaptive verification layers, and government-backed digital ID frameworks will emerge to prioritize ongoing authentication instead of one-time proof.&nbsp;</p><p><strong>Tip: </strong>Until then, individuals should share ID documents only through official websites or apps they have navigated to directly. Avoid uploading credentials through unsolicited links, and enable transaction alerts or credit freezes whenever possible to detect and stop fraudulent activity early.<strong>&nbsp;</strong>&nbsp;</p><h2>5. The browser becomes the new ground zero for deception&nbsp;</h2><p>By 2026, the browser will no longer be a neutral window into the internet. It will become the primary arena where criminals launch, automate, and scale their attacks. As people rely on a small handful of tabs to shop, bank, work, and communicate, attackers are moving directly into that space and using AI to blend in.&nbsp;</p><p>Malvertising, search ads, and sponsored posts will evolve from a nuisance to a mainstream threat. AI-generated ads will mimic brand visuals, copywriting styles, and product listings with near-perfect accuracy, allowing attackers to place fake links above legitimate ones. Clicking what appears to be your bank, a package delivery service, or a well-known retailer can now lead to a cloned site designed to harvest credentials or payment information. Entire counterfeit storefronts will be created in seconds, complete with AI-generated product images, scripted chat support, and fake tracking numbers that vanish once payment is processed.&nbsp;</p><p>Inside the browser itself, malicious scripts, fake update prompts, poisoned pop-ups, and session token theft will quietly replace older tactics that relied on downloads. Malware will increasingly live inside the page. Attackers will steal authentication tokens that keep people logged in, giving them direct access to accounts without ever needing a password. Even cautious users can be fooled when the entire environment around them looks legitimate.&nbsp;</p><p><strong>Tip:</strong> Combine strong browser hygiene with careful shopping habits. Enable passkeys or two-factor authentication for important accounts and regularly review active sessions. Use bookmarks or manually typed addresses for banking and government services instead of clicking ads or sponsored results. When shopping, treat unfamiliar online stores like physical ones. Look for real contact information, clear return policies, and secure payment options. If anything feels vague or automated, run the site through a security product or scam detector before entering payment details. Finally, choose a browser that prioritizes safety and privacy such as Norton Neo, the first AI browser built to be both safe and intelligent.&nbsp;&nbsp;</p><h2>The bottom line: 2026 is the year digital safety becomes personal&nbsp;</h2><p>The threats of 2026 are no longer abstract cyber risks happening “out there.” They are intimate, adaptive, and designed to meet people where they live online – on phones, in inboxes, in search results, and inside the browser tabs we trust most.&nbsp;</p><p>This is the year everyday people must build new reflexes.&nbsp;<br>This is the year organizations must rethink identity and authenticity.&nbsp;<br>This is the year technology companies must prioritize verification, not convenience.&nbsp;<br>And this is the year society must finally accept that safety is a shared responsibility.&nbsp;</p><p>The good news: awareness is power. These predictions aren’t just warnings; they are a roadmap. They show where attackers are headed and where defenses must evolve. With the right habits, tools, and skepticism, individuals can stay ahead of even the most sophisticated AI-driven threats.&nbsp;</p><p>Because while technology may be changing faster than ever, one truth remains: the more people understand the landscape, the harder it becomes for cybercriminals to win.&nbsp;</p>

As AI accelerates, identity, truth and trust face their biggest test yet. Here’s what to expect in 2026.

Cyber predictions for 2026

<p>The holiday season should feel festive, and not stressful. But every year, as shopping carts fill up and inboxes overflow with deals, scammers see an opportunity to strike. They know people are distracted, tired, and moving fast. That’s why some of the most convincing fraud attempts of the year happen between November and January.</p><p>At Gen, we work every day to help people stay ahead of evolving threats. And during the holidays, the smallest moments, buying a gift card, donating to a charity, or opening a “pre-approved” offer can carry outsized risk. Here’s what to watch for and how to protect yourself, based on the latest insights and practical guidance from our teams and partners from the Identity Theft Resource Center.&nbsp;</p><h2>1. Gift cards: convenient for you, gold for scammers</h2><p>Gift cards remain one of the easiest tools for fraudsters to exploit. They’re fast, unregulated, and nearly impossible to trace once the numbers are handed over. If anyone pressures you to pay for <em>anything</em> with a gift card, a fee, a fine, a service, or a “problem with your account,” that’s a scam.</p><p>Before you buy:</p><ul><li>Purchase cards only from reputable retailers.</li><li>Never share the numbers or PIN with someone you don’t know.</li><li>Ignore any links sent to “activate,” “verify,” or “unlock” a card; scammers increasingly use fake branded sites to steal balances.&nbsp;</li></ul><h2>2. Slow down before you donate</h2><p>Scammers build fake charities designed to look almost identical to the real thing, especially during high-giving months. They count on people acting quickly and emotionally.</p><p>Before donating:</p><ul><li>Search for the organization directly, not through a link in an email or text.</li><li>Check the name carefully. Misspellings, extra hyphens, or odd URLs are red flags.</li><li>Verify legitimacy through trusted sites like Charity Navigator or the IRS database.&nbsp;</li></ul><p>Real charities don’t rush you. If you feel pressured, step back.</p><h2>3. “Special offers” that want your SSN? Hard pass.</h2><p>Fraudsters love hiding behind fake credit card offers and holiday promotions. These scams are designed to harvest personal and financial information before you even realize what’s happening.</p><p>Keep yourself safe by:</p><ul><li>Avoiding offers that ask for full SSNs or bank details upfront.</li><li>Checking for HTTPS and secure branding before entering sensitive info.</li><li>Monitoring your credit for any newly opened accounts or inquiries you didn’t recognize or authorize.&nbsp;</li></ul><p>If an offer looks too shiny, too generous, or too urgent, trust your instincts.</p><h2>4. Your best defense: layers of strong, everyday protection</h2><p>The smartest way to stay safe isn’t about memorizing scam techniques. It’s about building habits that work year-round. As <a href="https://cmsb.nortonlifelock.com/sites/default/files/2025-12/Keeping%20You%20and%20Your%20Money%20Safe%20During%20the%20Holiday%20Crunch-2.pdf" target="_blank" title="Tips for Holiday Safety">this guide </a>emphasizes, your strongest protection comes from awareness paired with simple safeguards.&nbsp;</p><p>Set yourself up for success by:</p><ul><li>Checking your accounts regularly for unfamiliar transactions.</li><li>Using strong, unique passwords for every site you shop or bank with.</li><li>Turning on two-factor authentication wherever possible.</li><li>Relying on trusted security tools that monitor for identity threats, malicious sites, and scam attempts in real time.</li></ul><p>These small steps make it dramatically harder for attackers to turn holiday pressure into holiday losses.</p><p><strong>Final thought: Give yourself the gift of a calmer season</strong></p><p>Staying safe online shouldn’t take the joy out of the holidays. With a little awareness and the right protections in place, you can shop, donate, and celebrate with confidence.</p><p>The holiday crunch is real. But so is your power to stay one step ahead.</p>

Holiday scam trends and expert advice to keep your finances and identity protected.

Stay smart this holiday season: How to keep your money and identity safe when scams spike

Threat Research Team

<p>At the start of 2025, we anticipated a turning point in cybersecurity. AI was accelerating, threats were evolving, and people were becoming increasingly entangled in a digital world that demanded new instincts and stronger defenses. Now, as the year comes to a close, it’s clear that many of the trends we foresaw have not only emerged but become defining realities for people and global security at large.&nbsp;</p><p>Below is a look back at each major prediction and what materialized in 2025.&nbsp;</p><h2>1. AI will blur the line between reality and fiction&nbsp;</h2><h3>What we predicted&nbsp;</h3><p>Artificial intelligence isn’t just a powerful tool; it’s becoming a force that reshapes how we perceive the world. But, as it becomes better, smarter, more realistic and accessible, AI will increasingly blur truth and deception, making it more difficult than ever to decipher what’s real.&nbsp;</p><h3>Hyper-personalized realities&nbsp;</h3><p>AI-powered Large Language Models (think ChatGPT, Gemini, etc.) are advancing at an incredibly fast pace, offering hyper-personalized interactions that could alter perceptions and decision-making.&nbsp;&nbsp;</p><p>Imagine a world where AI doesn’t just answer your questions but anticipates your needs.&nbsp;</p><p>But here’s the trade-off: to unlock this seamless, personal assistant experience, people are handing over everything to AI, including calendars, messages and apps. Hyper-personalization comes with a cost: the erosion of privacy and autonomy. Are you willing to lose privacy for convenience?&nbsp;&nbsp;</p><p>This raises critical ethical questions: Who controls the narratives AI presents? How do we safeguard independent thought in a world increasingly influenced by machine intelligence?&nbsp;</p><p>We believe that AI-driven solutions should ensure transparency and accountability, giving individuals tools to navigate these hyper-realities without losing sight of two key things: truth and trust.&nbsp;</p><h3>AI in sensitive roles&nbsp;</h3><p>We suspect that AI will also infiltrate areas once considered uniquely human, such as parenting and education. Imagine an AI tool stepping in to mentor a child, help with homework, or even act as a digital babysitter. While it offers unlimited time and patience, it challenges the boundaries of what roles and responsibilities should be left to humans.&nbsp;</p><h3>Redefining relationships</h3><p>AI is poised to transform the family, from parenting and pets to companionship, by taking over routine tasks like homework help and schedule management or even emotional connections. Bonding moments – like bedtime stories or daily conversations – slip away. With robotics bringing AI into the physical world, the stakes grow higher: robotic pets replacing real ones, AI nannies soothing children and robot companions filling emotional gaps.&nbsp;&nbsp;</p><p>Shaping young minds: As AI becomes more engaging, children may form stronger bonds with machines than caregivers, shifting family dynamics. While AI offers personalized guidance and companionship, overdependence could limit kids’ social and emotional growth, making real relationships harder to navigate. Families must balance the benefits of AI with the need for genuine human connection.&nbsp;</p><p>The broader societal question is this: how much influence will we allow AI to have in shaping the next generation?&nbsp;</p><h3>What came true</h3><p>This prediction came true, but materialized even faster than expected. Throughout 2025, AI-generated images, audio and written content reached a level of realism that fooled not only consumers but also experienced content moderators and security teams. Several high-profile incidents involved deepfake audio used to authorize fraudulent wire transfers, and social platforms struggled to contain synthetic celebrity scandals that spread faster than verified corrections. The year proved that users need stronger tools and instincts to distinguish real from manufactured content.&nbsp;</p><p>AI in sensitive roles became one of the most unexpected and concerning shifts of the year. Families increasingly leaned on AI for tutoring, companionship and parenting support, but the biggest wake-up call came from children using ChatGPT-style tools for therapy. Schools and parents reported kids turning to AI to talk through anxiety, friendship problems and even serious emotional issues before speaking to adults. What started as “homework help” quietly evolved into emotional outsourcing, raising big questions about dependency, privacy and the boundaries between guidance and guardianship. In fact, based on our own consumer survey from summer 2025, <a href="https://newsroom.gendigital.com/Norton-Cyber-Safety-Insights-Report-Connected-Kids" target="_blank" rel="noreferrer noopener">NCSIR: Connected Kids</a>, over a third of parents say their child uses AI for emotional support, a number which rises to 41% in the United States. The prediction that AI would move into roles once considered uniquely human proved not only accurate, but urgent.&nbsp;</p><h2>2. The deepfake revolution&nbsp;</h2><h3>What we predicted</h3><p>Deepfakes, AI-generated media designed to mimic real people, are becoming so sophisticated that even experts may struggle to distinguish truth from fabrication. In 2025, we expect increases in:&nbsp;</p><h3>Personal attacks</h3><p>Scorned individuals could use deepfakes to harass or extort others, not just by targeting the individual directly but by creating convincing fake media of family members or friends. These fabricated videos or audio clips could be used to manipulate victims emotionally, spread false rumors, or even strain personal relationships, amplifying the psychological toll of such attacks.&nbsp;</p><h3>Political manipulation</h3><p>Governments and bad actors may leverage deepfakes to divide, spread disinformation and destabilize societies. Imagine a fake speech from a world leader announcing false policies — such content could ignite panic, erode trust in institutions and manipulate public opinion. Deepfakes could also target journalists or political opponents, fabricating damaging scandals or discrediting reliable sources, undermining public confidence in the media. This highlights the need for a system where authenticity is non-negotiable, where digital signatures could become the standard to restore trust.&nbsp;&nbsp;</p><h3>Financial fraud</h3><p>Deepfake videos or audio clips of executives could be used to impersonate authority figures, convince employees to transfer funds, reveal sensitive company information or approve unauthorized transactions. For instance, a Business Communication Compromise (BCC) can happen when a convincing video of a CEO instructing a financial officer to expedite a payment bypasses typical safeguards, due to its perceived authenticity and established trust. The threat doesn’t end with internal fraud; deepfakes could also target investors or clients, undermining trust in corporate communications and causing long-term reputational damage. These attacks could ripple through supply chains and financial systems, creating chaos on an unprecedented scale.&nbsp;</p><h3>What came true&nbsp;</h3><p>Deepfake-driven incidents exploded in 2025. Several political events were disrupted by fabricated videos released ahead of elections, forcing governments to fast-track authenticity verification standards. On the personal side, extortion attempts using manipulated media targeting both adults and teens rose sharply. Corporate deepfake fraud also became more frequent, with attackers impersonating executives via video to approve payments or release confidential information. The need for digital signatures and content provenance became a critical industry focus.&nbsp;</p><h2>3. Data theft: the multi-faceted threat to identity&nbsp;</h2><p>The digital world runs on data and cybercriminals are mastering the art of leveraging it as a multifaceted threat to target individuals, organizations and societies alike.&nbsp;</p><h3>Comprehensive profiles for targeted scams&nbsp;</h3><p>Large-scale breaches and public data sources provide criminals with the raw materials to build highly detailed profiles of their victims.&nbsp; This goes beyond just a first and last name or email address – criminals piece together “who” their targets really are. They know where you work, what you do for a living, your hobbies and habits. With this intimate understanding and detailed information, they can:&nbsp;</p><ul><li data-list-item-id="e89b802767e6107cd5138f2be84805b96">Craft hyper-personalized phishing attempts</li><li data-list-item-id="ecc1dc1f1d32ce29099862abfb9c8c717">Launch convincing extortion schemes</li><li data-list-item-id="eef6e50d0563476d6c1c7a0871d7f86b2">Impersonate trusted services to exploit victims’ trust&nbsp;</li></ul><p>For example, attackers using images of victims’ homes during sextortion attempts demonstrate how cybercriminals are evolving their psychological manipulation tactics.&nbsp;</p><h3>Forgotten logins, fresh targets: how old Accounts open new doors for hackers&nbsp;</h3><p>The average individual has hundreds of online accounts. Besides the active ones, there’s also a plethora of aged accounts that are long gone, such as old email addresses from Hotmail or social accounts like MySpace. These accounts can be forgotten or abandoned. Combined with the low adoption of multi-factor authentication (MFA) and the widespread practice of password reuse, these neglected accounts become gateways for attackers. Cybercriminals can pivot from one breached account to another, collecting more personal information with each step and enabling even more precise and damaging attacks.&nbsp;</p><h3>What came true</h3><p>This prediction proved to be one of the year’s most accurate. Cybercriminals leaned heavily on deep, blended data sets to create unnervingly complete profiles of their targets. Sextortion cases using photos pulled from social media and real home exterior images surged. Hyper-personalized phishing hit new levels of realism as attackers referenced victims’ workplaces, travel plans, hobbies and even recent online purchases. And perhaps most striking, the “forgotten logins” problem exploded: millions of abandoned Hotmail, Yahoo and MySpace-era accounts were compromised and used as stepping stones into more active identities. Once inside, attackers chained together old passwords, reused credentials and unsecured inboxes to pivot across services, proving that yesterday’s accounts are today’s attack surface.&nbsp;</p><h2>&nbsp;4. Scams enter the era of hyper-personalization&nbsp;</h2><h3>What we predicted</h3><p>2025 marks the dawn of hyper-personalized scams. Cybercriminals no longer rely on broad attacks and wait for someone to take the bait. Instead, they use meticulously crafted profiles built from breached data, public records or dark web scraps to tailor their attacks. They’re not random. They’re designed to feel personal, disarming and hard to resist because they rely on your specific data – your own identity.&nbsp;&nbsp;</p><p>These scams won’t feel like scams. They’ll feel real. Here are some examples:&nbsp;</p><h3>Tailored manipulation&nbsp;</h3><p>Imagine receiving a message referencing a conversation you had last week or mimicking a friend’s tone perfectly. Attackers will use personal details to craft scams so convincing they disarm even the most vigilant.&nbsp;</p><h3>Psychological exploits</h3><p>Cognitive biases like urgency, trust, and fear will be turned against us. Emotional triggers — like a fabricated crisis involving a loved one — will make hesitation feel impossible.&nbsp;</p><h3>Platform integration</h3><p>Social media and messaging apps will become prime battlegrounds, with scams that look indistinguishable from legitimate interactions.&nbsp;</p><p>This is the reality of where scams are heading. The lines between genuine and fraudulent will blur, forcing us to rethink how we defend against threats. Technology together with awareness, education, and proactive resilience will be critical to protecting ourselves and our communities in this new era of deception.&nbsp;</p><h3>What came true</h3><p>The shift to hyper-personalized scams was unmistakable. Fraudsters used real-time AI to mirror the writing style, vocabulary and emotional tone of friends, family members and trusted companies. Consumers reported receiving messages referencing actual conversations, locations and past purchases. Social platforms became major battlegrounds for highly tailored scams that felt indistinguishable from legitimate interactions. This was the year when “it felt too real to be a scam” became a recurring theme in victim reports.&nbsp;</p><h2>5. Financial theft: the new frontier of exploitation&nbsp;</h2><h3>What we predicted</h3><p>The war on financial security is escalating, with attackers blurring the boundaries between digital fraud and physical coercion. 2025 will see financial theft reach unprecedented levels, fueled by innovation in both technology and tactics.&nbsp;</p><h3>A new breed of digital fraud&nbsp;</h3><p>Cybercriminals are wielding sophisticated tools to undermine trust and exploit vulnerabilities:&nbsp;</p><h3>Deepfake scams&nbsp;</h3><p>Picture a video of a trusted leader or celebrity endorsing a high-return investment. These convincing forgeries will draw in victims by the thousands, fueling a new wave of financial deception.&nbsp;</p><h3>Voice-cloned lies</h3><p>Fraudsters will impersonate government officials, announcing fake income distributions or policy updates, driving victims to malicious platforms.&nbsp;</p><h3>Crypto scams on steroids</h3><p>From fake giveaways to fabricated trading platforms, cryptocurrency will remain a prime target for attackers hungry for unregulated, high-reward opportunities.&nbsp;</p><h3>Where cybercrime gets personal&nbsp;&nbsp;</h3><p>Financial theft is no longer confined to just the digital world. Now, it feels more and more personal. Spyware and malware will silently monitor devices, giving attackers backdoor access to banking apps and bypassing traditional security measures.&nbsp;</p><p>This evolving threat challenges us to rethink financial security. It’s not just about protecting devices; it’s about protecting the people behind them. The only way forward is bold innovation, collective vigilance, and an unwavering commitment to staying ahead of those who prey on trust.&nbsp;</p><h2>Our vision for the future&nbsp;</h2><p>The challenges of 2025 demand bold action, innovative thinking and a commitment to protecting what matters most: trust. And while threats are undoubtedly getting more complex, the future is bright, and technology will play a pivotal role in making the world safer, smarter and more secure. At Gen, we’re not just observing these challenges; we’re staying ahead of them. With products like Norton Genie for scam detection, Norton 360 Deluxe and Avast One for all-in-one protection, and LifeLock to restore what’s lost, we’re safeguarding your digital world and Powering Digital Freedom.&nbsp;</p><p>While technology like AI holds the promise of a more secure and innovative future, uncertainty often stands in the way of progress. This is where we see opportunity: to educate, empower and lead. By helping others understand and embrace AI responsibly, we can ensure it becomes a force for good — strengthening our defenses while driving innovation.&nbsp;</p><p>Make no mistake. New technology breeds new risks. However, it also serves as a powerful catalyst for positive change, paving the way to a safer and brighter future.&nbsp;</p><p>With every new tool, every evolving threat, and every transformative idea, we’re shaping a future where innovation and security go hand in hand.&nbsp;</p><h3>What came true</h3><p>The financial sector experienced one of its most turbulent years. Deepfake investment pitches, AI-generated endorsements and cryptocurrency fraud campaigns drew in record numbers of victims. Malware designed to capture session tokens from banking apps became more common, allowing criminals to bypass passwords altogether. We also saw a rise in hybrid threats where digital scams led to in-person coercion or targeted follow-ups. Financial institutions accelerated their move toward biometric verification and real-time behavioral fraud detection as a result. &nbsp;</p><h2>Looking ahead&nbsp;</h2><p>Our <a href="https://www.gendigital.com/blog/insights/leadership-perspectives/predictions-2025" title="2025 Gen Cybersecurity Predictions">2025 predictions</a> anticipated major shifts across AI, identity, scams and financial crime. What we saw this year confirmed that these transformations are not abstract forecasts but active forces shaping daily life. The pace of change has only accelerated, which is why our focus in 2026 will be on helping people build new digital instincts, adopt stronger security tools and navigate an internet where deception is easier than ever to scale. And with the landscape shifting again, stay tuned, our 2026 predictions are on the horizon.&nbsp;</p>