Global Head of Scam Research

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

Corporate Manager

Leadership Perspectives

There’s a quiet shift happening with money right now. 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.So let’s strip it back. Financial safety isn’t about being perfect. It’s about being prepared before something goes sideways.&nbsp;<h4>Start with this: safety beats strategy</h4>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.But here’s the part that gets skipped: none of that works if your foundation isn’t solid.Financial safety starts with three basics: • Cash buffer (even a small one) • Controlled spending • Awareness of risk (scams, debt, bad advice)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.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.But it does mean one thing: building your own safety net matters more than ever.&nbsp;<h4>The smartest move: know where your money is going</h4>You don’t need a complicated system. You just need visibility.Call it budgeting, call it tracking, call it not being surprised at the end of the month. It’s all the same thing.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.Translation: You’re not broke. You’re choosing where your money goes.That mindset shift alone is powerful.&nbsp;<h4>Build a buffer before you build wealth</h4>It’s tempting to jump straight into investing. But if one unexpected expense can wipe you out, you’re building on shaky ground.Start smaller: • Aim for $500 to $1,000 in emergency savings • Then grow toward 3 to 6 months of expensesIt doesn’t happen overnight. That’s fine. Consistency beats intensity here.Because when something goes wrong, and it will, you won’t have to rely on high interest debt to survive it.&nbsp;<h4>Watch the digital traps</h4>This is where financial safety gets very real.Gen Z lives online. That means your money does too. And that’s exactly where scammers operate.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.And it works.There’s also a gap worth calling out: confidence doesn’t always equal protection.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>)And when scams work, they don’t just take money. They take identity.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>)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.As Eva Velasquez, CEO of the Identity Theft Resource Center (ITRC), puts it: “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.”If your financial life lives online, your identity is part of your financial safety net. Protecting it isn’t optional anymore. It is foundational.A few rules that hold up:<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>Being skeptical isn’t negativity. It’s protection.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.

<h4>Use tools that actually help you move forward</h4>This part matters more than ever: you don’t have to figure all of this out alone.The line between financial health and digital safety is disappearing. That’s why platforms are starting to bring both together.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.Because financial safety isn’t just about avoiding risk. It’s about having the confidence to act.&nbsp;<h4>You’re not behind. You’re early</h4>It’s easy to feel like you’re late with money. Like you should already have it figured out.You’re not. You’re early.Gen Z is already reshaping how money works. There are more open conversations, earlier investing and more willingness to question the system.The goal isn’t perfection. It’s momentum.Start with:<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>That’s Financial Safety 101.Not flashy. Not viral. But it works.And once that foundation is there, everything else, investing, growing, building real wealth, gets a whole lot easier.

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

Security Evangelist at Gen

“Is this a scam?” has become one of the defining questions of the digital era.&nbsp;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;At Gen, we see this as more than a trend. It signals the emergence of a new trust architecture for the internet.&nbsp;<h2>AI is becoming the first line of defense&nbsp;</h2>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;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;That convergence creates both opportunity and responsibility. If AI is where decisions are being shaped, then security intelligence must live there too.&nbsp;<h2>From reactive security to decision-centric protection&nbsp;</h2>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;But this product moment sits within a much larger transformation.&nbsp;<h2>The rise of autonomous AI and the trust imperative&nbsp;</h2>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;Autonomy without guardrails amplifies risk. Autonomy with transparency amplifies trust.&nbsp;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;<h2>Designing AI that augments human judgment&nbsp;</h2>The future of AI interaction is not about replacing decision-making. It is about strengthening it.&nbsp;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;That balance between automation and empowerment is critical.&nbsp;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;<h2>Building the next trust layer of the internet&nbsp;</h2>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;The real question is whether we build it with trust at the core.&nbsp;At Gen, our approach is grounded in three ideas:&nbsp;<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>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;“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;The future of cybersecurity will be defined by how well we design trust into every layer of AI interaction.&nbsp;And that future is being built now.&nbsp;

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

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;&nbsp;Agentic AI is about to change that ratio. Not by replacing researchers, but by shifting where human judgment actually matters.&nbsp;<h5>&nbsp;The Work Today&nbsp;</h5>&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;&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;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;That's where agents come in.&nbsp;<h5>Agents as a Junior Researcher &nbsp;</h5>The useful mental model isn't "AI replaces researchers." It's "AI becomes the junior researcher you never had enough of."&nbsp;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;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;The craft doesn't disappear. It moves up the stack.&nbsp;<h5>The Real Opportunity: Scope&nbsp;</h5>Speed is the obvious benefit. But the real transformation is scope.&nbsp;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;When the grind gets automated, you can finally ask bigger questions:&nbsp;<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>Agents don't just make you faster. They make exploration possible in ways it wasn't before.&nbsp;<h5>The Other Side of the Equation&nbsp;</h5>Here's the part we have to be honest about: attackers get agents too.&nbsp;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;This isn't a one-sided advantage for defenders. It's an escalation. The ceiling rises for everyone. &nbsp;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;<h5>What Changes for Researchers&nbsp;</h5>The researchers who thrive in this shift will share a few traits:&nbsp;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;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;Adaptability. The tools will keep changing. The researchers who stay effective will be the ones who continuously experiment.<h5>The Craft Evolves&nbsp;</h5>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;I don't see it that way.&nbsp;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;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;That's not a loss. That's what progress in a field looks like.&nbsp;

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 

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.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.But as technology reshapes how we meet and connect, it’s also reshaping how trust can be manipulated.The latest Norton Insights Report,&nbsp;<a href="https://us.norton.com/blog/research/romance-scams-insights-report-2026">Artificial Intimacy</a>, reveals just how deeply artificial intelligence is transforming digital intimacy — and how scammers are adapting just as quickly.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.<h2>What are AI romance scams?</h2>AI romance scams are online dating scams that use artificial intelligence to create convincing fake personas and emotionally persuasive conversations.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.The result: deception that feels tailored, attentive and very real.<h2>How is AI changing online dating and digital trust?</h2>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.At the same time, 81% of respondents report experiencing loneliness, with younger generations reporting even higher levels.This signals something bigger than curiosity. It reflects how normalized AI-mediated emotional connection has become.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.That shift changes how emotional bonds are built online. And scammers are paying attention.<h2>How Do AI Romance Scams Work?</h2>Romance scams follow a predictable pattern:<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>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.Globally, romance scams already account for billions in losses each year. As AI improves realism and scalability, the risk compounds.What makes 2026 different is the sophistication. AI allows scammers to:<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>Even with advanced tools, warning signs still exist. Red flags may include:<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>If emotional intensity rises faster than real-world verification, it’s time to pause.

<h2>Why do romance scams increase around Valentine’s Day?</h2>Valentine’s Day amplifies emotional vulnerability. More people are actively dating, messaging and seeking companionship. That surge in activity creates opportunity.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.“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 trustWhen loneliness intersects with seasonal pressure and AI-powered persuasion, risk increases.<h2>How can you protect yourself from romance scams?</h2>Awareness remains one of the strongest defenses.Practical safeguards include:<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>Healthy skepticism does not mean rejecting connection. It means layering trust with verification.<h2>Can you fall in love with AI?</h2>The data suggests many people believe it’s possible.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.

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

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.<h2>What is OpenClaw and what does it do?</h2>OpenClaw (formerly known as Clawdbot and Moltbot) is an open-source autonomous AI agent 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;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 skill&nbsp;— basically a set of instructions that tells OpenClaw what to do and how to do it.&nbsp;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.But OpenClaw goes far beyond prepackaged features: it can also write its own skills. When prompted by a user, OpenClaw can create&nbsp;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.

<h2>OpenClaw is transformative, but is it secure?</h2>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>.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.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.<h3>How AI agents are changing the rules of security</h3>Using an AI agent means delegating responsibility 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;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:<ul type="disc"><li data-list-item-id="e4fca606bedbd6d9f48f918a03f9af53f">Access to private data: The agent can see your files, login credentials, and personal information.&nbsp;</li><li data-list-item-id="ecf3a565dd59d91bd98732e5db397a3b0">Exposure to untrusted content: The agent can process and interact with messages, emails, websites, prompts, and third-party integrations.</li><li data-list-item-id="eecca7e258f9c3ab0e8ef4aa8bfdfcb46">The ability to communicate externally: The agent can send messages, call APIs, and run commands on your behalf.</li></ul>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.<h2>What are the core security risks of OpenClaw?</h2>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;Here are some of the main risks that AI agents like OpenClaw introduce:<ul type="disc"><li data-list-item-id="e7f1f8c06c1fb4bba4ae7ae46658102f6">Larger attack surface: 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">Prompt injection attacks: 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">Malicious skills and plugins:&nbsp;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">System-wide impact: When OpenClaw runs with extended permissions, mistakes or compromises can affect your entire system.</li><li data-list-item-id="e45a537af168a7d6ebe7de9cfd9c41209">Illusion of judgment: 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>Gen Threat Labs has identified OpenClaw skills as one of the&nbsp;first large-scale threat vectors in autonomous AI agents built for consumers.&nbsp;In a recent analysis, Gen security researchers discovered that nearly 15% of OpenClaw skills contained malicious instructions, 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.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.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;The OpenClaw skills risk that Gen researchers have identified is called delegated compromise, which is when an attack targets an agent and thereby exploits all the permissions the agent has been given.&nbsp;<h2>How to use OpenClaw more safely</h2>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;If you want to experiment with OpenClaw, here’s how to do it more safely:<ul type="disc"><li data-list-item-id="e908f0fbadded99d4c5344d6f31eee365">Run OpenClaw in a secure environment: 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">Don’t connect OpenClaw to the internet: 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">Start with limited permissions and expand slowly: 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">Don’t connect sensitive accounts: 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">Keep your human judgement at the center: 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">Use safe communication channels: 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">Be careful with third-party skills and plugins: 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">Scan new skills before installing them: 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>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.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;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.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.

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

As spring blooms and wedding season approaches, love is in the air — and in dating apps. Singles are swiping, chatting and hoping to find something real. But in a digital world where DMs replace first dates and algorithms spark connections, scammers are circling too.&nbsp;<a href="https://www.aarp.org/money/scams-fraud/what-is-catfishing.html%22%20/h%EF%B7%9FHYPERLINK%20%22https://us.norton.com/blog/online-scams/romance-scams" target="_blank" rel="noreferrer noopener">Romance scams</a> aren’t new. But fueled by artificial intelligence, they’re no longer just deceitful — they’re calculated, scalable and deeply personal. What once took hours of emotional manipulation can now be automated, translated and tailored in seconds.&nbsp;AI hasn’t just changed how these crimes work. But, it has changed its reach.&nbsp;At Gen, we refuse to accept this as the cost of connection.&nbsp;<h2>The AI scam surge</h2>According to the latest <a href="https://newsroom.gendigital.com/norton-cyber-safety-report-2025" target="_blank" rel="noreferrer noopener">Norton Cyber Safety Insights Report</a>, two in five people on dating apps report being targeted by scams. Of those, 41% have fallen victim. And our research with <a href="https://us.norton.com/products/genie-scam-detector" target="_blank" rel="noreferrer noopener">Norton Genie</a> (our AI-powered scam detection tool and assistant) revealed an even more chilling trend: 60% of online daters believe they’ve chatted with someone whose messages were AI-generated.&nbsp;When it comes to the types of scams online daters in the U.S. are facing, here's what they report: &nbsp;<ul><li>55% have encountered catfishing</li><li>34% have experienced romance scams</li><li>32% have been targeted by sugar daddy or sugar baby scams</li><li>31% have faced sextortion scams</li></ul>These numbers underscore just how widespread and varied scam tactics have become, preying on vulnerability, trust and the desire to connect.In the U.S., we see that Congress is&nbsp;considering new measures to combat the rise of romance scams with the introduction of the Romance Scam Prevention Act (<a href="https://www.govtrack.us/congress/bills/119/s841" target="_blank" rel="noreferrer noopener">S.841</a> and <a href="https://www.govtrack.us/congress/bills/119/hr2481" target="_blank" rel="noreferrer noopener">H.R. 2481</a>). If passed, it would:&nbsp;<ol><li>Require platforms to notify users exposed to flagged profiles</li><li>Mandate investigations and suspensions of suspicious accounts within 24 hours</li><li>Provide scam awareness resources directly to users&nbsp;</li></ol>The proposed legislation follows previous efforts such as the <a href="https://www.govtrack.us/congress/bills/118/hr6125" target="_blank" rel="noreferrer noopener">Online Dating Safety Act</a> of 2023. The latest proposal includes expanded provisions and would authorize enforcement by the Federal Trade Commission (FTC), with violations subject to federal action.&nbsp;<h2>Our role: Tech must lead, not lag&nbsp;</h2>At Gen, we try not to wait for mandates. We would rather lead with real-life solutions.&nbsp;<ul><li>We built Norton Genie to give people real-time answers when something feels off, no matter where that conversation started.</li><li>We’ve conducted deep research into AI-enabled romance scams to understand how they work and evolve.</li><li>And we believe platforms need education, transparency and user protection.&nbsp;</li></ul>As romance scams evolve with AI, we must evolve our approach to protect consumers through smarter tools, better education and collaborative policy solutions.&nbsp;&nbsp;<h2>This fight belongs to all of us&nbsp;</h2>This is not just a dating app problem. Romance scams thrive across texting platforms, email, gaming sites and social media feeds. Basically, anywhere people connect.&nbsp;That’s why Gen joined the <a href="https://www.gasa.org/" target="_blank" rel="noreferrer noopener">Global Anti-Scam Alliance</a> (GASA). Because protecting people from digital deception takes collaboration, not competition. Consumer safety is not a partisan issue — it’s a human one. We’re committed to empowering consumers by conducting in-depth scam research, educating users about potential threats, and working alongside policymakers and industry partners to better protect people from scams.&nbsp;We believe in a future where technology empowers people, not preys on them. A future where we can connect without fear. Where love is honest, not engineered by a bot. Where trust is earned — and protected — by all of us.&nbsp;&nbsp;&nbsp;Read more about our <a href="https://www.gendigital.com/blog/insights/leadership-perspectives/ai-2025-gen-policy" target="_blank" rel="noreferrer noopener">policy recommendations</a>.&nbsp;&nbsp;&nbsp;

Why we’re fighting back against AI-driven romance scams

Love in the age of algorithms 

We leverage the strength of our global team, knowing we’re more powerful together. Read news about Gen's company activities and initiatives.

Company News

Gen has a family of trusted consumer brands that for decades have helped people stay safe in the digital world. We pride ourselves in the trust we’ve built with people around the world to navigate the latest technology safely and with confidence. And that’s exactly why we recently launched the Gen Agent Trust Hub – our trust layer for the agentic AI ecosystem.AI agents are no longer theoretical. They are here. They are editing files, installing packages, calling APIs, connecting to MCP servers, spawning sub-agents, persisting memory, and executing real workflows across environments. This shift from passive AI to autonomous execution fundamentally expands the attack surface.The Agent Trust Hub (ATH) exists to meet this moment. It is designed to provide continuous protection across the full lifecycle of an AI agent – from identity and verification to runtime enforcement. But security at this scale cannot rely on proprietary guardrails alone. It requires shared foundations the industry can build on.To help establish that foundation, we created the AI Agent Safety Standards by Gen to serve as a unified framework designed to bring consistency, portability, and accountability to agentic systems. This framework is currently built on two complementary pillars:<ul type="disc"><li data-list-item-id="e464ce0354d35c1a986c83945bc06b0cd">AI Agent Runtime Safety Standard (AARTS) – an open standard for runtime decision enforcement across agent hosts</li><li data-list-item-id="e4a7d4ceb7039019b7c2554401a8bb1c3">Skill IDs – a deterministic, content-addressable fingerprinting system for AI agent skills</li></ul>Together, they enable the ATH and the broader agentic AI community to enforce trust not just within one environment but across the broader agentic ecosystem. &nbsp;Let’s dig into the details.&nbsp;<h2>Pillar One: Standardizing Runtime Safety for AI Agents</h2><h3>AARTS</h3>AI agents operate across a fragmented ecosystem of IDEs, orchestrators, frameworks, and standalone applications. Each host defines its own lifecycle events, security hooks, and enforcement logic.This fragmentation creates blind spots. A security engine integrated into one host may lack context or control in another, even when the risks are identical.AARTS&nbsp;addresses this gap. It is not a product. It is a vendor-neutral contract that defines:<ul type="disc"><li data-list-item-id="e226fc797549b56b0ae0a537923fd79f6">Where security decisions can be made (hook points)</li><li data-list-item-id="e7566ceb595a0278da9453bbd84659578">What data is available for evaluation (data model)</li><li data-list-item-id="e637f36bcc4e8dcfac7321ad59230c1c1">How decisions are enforced (verdict semantics)</li></ul>Simply put:&nbsp; AARTS standardizes runtime security decision-making across agent hosts.<h3>The Architecture</h3>AARTS defines three cleanly separated roles:<ol><li data-list-item-id="e7a445b9a896b9b89299afa17ca58c69e">Host Runs the agent, emits lifecycle events, and enforces decisions.</li><li data-list-item-id="e989b722cd161ff069c924d70571cd6b7">Adapter Maps host-native events into a standardized AARTS schema.</li><li data-list-item-id="ed862f4398b4b82c3b52c721801728a9c">Security Engine Evaluates events and returns structured verdicts.</li></ol>This separation prevents superficial “checkbox security” while preserving portability. Hosts provide the right context at the right time. Engines make decisions without depending on host internals.In ATH, this architecture allows enforcement logic to remain portable while enabling hosts to integrate in a predictable way.<h3>Hook Points: Where Safety Happens</h3>AARTS v0.1 defines 19 hook points across the agent lifecycle, including:<ul type="disc"><li data-list-item-id="ec5255705352d810833f77d79b310a42d">PreToolUse — evaluate shell commands, file writes, web requests, package installs</li><li data-list-item-id="ee1ebbdf3e4522a59dedaa55c0305d903">PreLLMRequest — protect prompt integrity and instruction layering</li><li data-list-item-id="e10ef155bfe4c584afd7492da267796be">PreSkillLoad / PrePluginLoad — enforce supply chain controls</li><li data-list-item-id="ee6184f1048dfbf4ee3c0e62c2d3dff16">PreMCPConnect — treat MCP connections as trust boundaries</li><li data-list-item-id="ebc57267bf0e891e01f9daf277ced1a01">PreMemoryRead / PreMemoryWrite — protect persistent memory</li></ul>These hooks enable consistent policy enforcement regardless of whether an agent runs in an IDE, CLI, or orchestrator.Verdict Semantics: Allow, Deny, AskAARTS engines return structured verdicts:<ul type="disc"><li class="ck-list-marker-bold" data-list-item-id="e9615d2d55e251652e73363c052de0bd1">Allow</li><li class="ck-list-marker-bold" data-list-item-id="eb510a0ee27b0a853b90ff01cec91953d">Deny</li><li class="ck-list-marker-bold" data-list-item-id="ed6e8c58ce99af62dddbf6b530bc0c83c">Ask</li></ul>Each verdict includes severity, threat category, optional confidence scoring, and human-readable reasoning.Importantly, AARTS enforces deterministic behavior:<ul type="disc"><li data-list-item-id="edfce7cb3a277e1da6e442c971f8cc8e1">If a host cannot support interactive “Ask,” it defaults safely to Deny</li><li data-list-item-id="ec790330b39c695d6cdacf4118e106c1b">If an engine fails, hosts fail-open and log explicitly</li></ul>These operational tradeoffs are intentional. Predictability is essential in security infrastructure.<h3>Why AARTS Matters</h3>Without a shared runtime contract:<ul type="disc"><li data-list-item-id="e2574817ec2bfc1a1b0233c5fca970994">Security integrations are bespoke and brittle</li><li data-list-item-id="ebbe5c354debe741cc08436719a34032e">Tool taxonomy is inconsistent</li><li data-list-item-id="e2f498a30e5d748e37c3176b169145707">Prompt injection controls vary by host</li><li data-list-item-id="e3aff5387909fd85f523c40998fdb13b0">Supply chain risk is unevenly enforced</li></ul>With AARTS:<ul type="disc"><li data-list-item-id="eaa8f402b20b740f0992dd0f9b4f42b44">One engine can operate across multiple hosts</li><li data-list-item-id="e3ddcda4fa65f1abd066be4b0bef0201b">Policies remain portable</li><li data-list-item-id="e5ec3076209374ae5eafb39a00f19adf4">Audit logs become comparable</li><li data-list-item-id="e6f17fa0cc0a6a6e0136b63569cdc58a6">Trust boundaries are explicit</li></ul>AARTS standardizes enforcement in the agentic world.To read the full technical breakdown, see the full AART&nbsp;<a href="https://www.gendigital.com/blog/news/company-news/ai-agent-runtime-security">technical blog post</a>. To comment on the standards, visit our page on&nbsp;<a href="https://github.com/gendigitalinc/aarts">GitHub</a>.<h2>Pillar Two: Skill IDs</h2><h3>Fingerprinting AI Agent Skills</h3>AI agents are extended through skills – directory bundles containing a SKILL.md prompt file alongside scripts, templates, configuration, and documentation.These skills function like plugins. Install one, and the agent gains new capabilities.But skills are not single files. They are directory trees.And directory trees are notoriously difficult to fingerprint reliably.<h3>The Identity Problem</h3>A naive SHA-256 hash of a ZIP file does not work:<ul type="disc"><li data-list-item-id="e6fc83ba16d130bd7d3d62ba2af79f96d">ZIP files are non-deterministic</li><li data-list-item-id="ee7939e0ca570cb480cbdc5b8e5460ece">Timestamps vary</li><li data-list-item-id="e5d398fbba0c41d53d0e3782902bce4c1">Entry order differs</li><li data-list-item-id="e95611c8e95e6ad4f16eccdede99ac42c">Unicode normalization differs across OS platforms</li><li data-list-item-id="ef8771bdef58d5438cc192edc36e89c56">Wrapper directory names vary</li></ul>Two identical skills can produce different ZIP hashes.That is unacceptable for security workflows.<h3>Designing the Skill ID</h3>Skill IDs solve this problem by creating a deterministic, content-addressable identifier for a skill’s logical directory tree.The algorithm:<ol><li data-list-item-id="ebdc8f7c1fc531f4ccbd79f2baf7b6d83">Extract all files and directories</li><li data-list-item-id="e06cf4d178c29a57575a249d22d5cb9b4">Normalize paths (slashes, Unicode NFC, dot components)</li><li data-list-item-id="e69f139d48a1c91815267ffb649c74e07">Strip wrapper directories</li><li data-list-item-id="e07c1da87b4b07a634d73a79ab93f9134">Hash each file individually (SHA-256)</li><li data-list-item-id="ef7cb735107df0cafdbcc921e639c0212">Build sorted tree entries using null-byte delimiters</li><li data-list-item-id="eab8b3782aab3b4dd9b0f4aed3255caf8">Hash the tree to produce the final Skill ID</li></ol>The result is a 64-character SHA-256 digest that changes if and only if meaningful content changes.Same content, different ZIP packaging? Same ID. One byte changed? Different ID.This mirrors git’s tree hashing model and decades of security best practices.<h3>What Skill IDs Enable</h3>A stable identity primitive unlocks established security workflows:Allowlisting Approved skills can bypass repeated analysis.Content-addressable storage Deduplicate identical submissions automatically.Per-file caching Reuse expensive AI analysis on unchanged files.Detection flagging Connect file hashes to reputation systems.Version tracking Content becomes the version boundary.Skill IDs provide identity before verdict.<h3>From Identity to Authenticity: Signing Skills</h3>Identity answers “what is this?”Authenticity answers “who published this?”We are exploring a signing model inspired by Authenticode:<ul type="disc"><li data-list-item-id="e2103c63753a0f24a87fa144d75611c7a">A skill.sig file embedded inside the directory</li><li data-list-item-id="e28e6643465e0cabeea54eb4bc5302028">Excluded from the Skill ID calculation</li><li data-list-item-id="e700fd96bf7f3d0db4aa70df0a92d57ec">Contains a signature over the Skill ID</li><li data-list-item-id="ea2483e766cae2defa9d2dd19707d9302">Verifiable against the author’s public key</li></ul>This enables:<ul type="disc"><li data-list-item-id="eddf73c0b79d7d482c030736b4f449be8">Author signing</li><li data-list-item-id="e3d708097518f6e38a31a73aaf27ecb49">Marketplace attestation</li><li data-list-item-id="e81ca5157252eece76405ff632c0d6722">Certificate chains</li><li data-list-item-id="efe336aa9f477f99f141fc5a81f887d9b">Revocation workflows</li></ul>Self-verifiable skills create accountability without external lookups.To read the full technical breakdown, see the full <a href="https://www.gendigital.com/blog/insights/research/ai-agent-skill-fingerprinting" target="_blank" rel="noreferrer noopener">SkillsID technical blog post</a>. To comment on the standards, visit our page on <a href="https://github.com/gendigitalinc/skill-id-standard">GitHub</a>.&nbsp;<h2>The AI Agent Safety Standards by Gen: Identity + Enforcement</h2>AARTS governs runtime enforcement. Skill IDs govern artifact identity.Together, they form the foundation of the AI Agent Safety Standards.<ul type="disc"><li data-list-item-id="ea3ef208939f34e4fcfaf98bd8e26b2d6">Skills can be uniquely identified, versioned, and signed.</li><li data-list-item-id="e30dceebd279e7a291dc71a82c7168ab2">Runtime actions can be evaluated consistently across hosts.</li><li data-list-item-id="e1a7b0d4259c4abb5e6ba2f38dffd6aaa">Supply chain trust and runtime enforcement operate under shared semantics.</li></ul>This is the same model that matured endpoint security over decades:<ul type="disc"><li data-list-item-id="ebacb70fb4a6565c103650a6c6ee67532">File hashes</li><li data-list-item-id="ed254dd1eec0c62e2aea6bce82e871995">Content signatures</li><li data-list-item-id="e37a3133a17b4a1adca017b868990baa6">Runtime enforcement</li><li data-list-item-id="eeeb5f5d7a84badb31f60600269d24edf">Portable policy engines</li></ul>We are applying those proven patterns to a new artifact type and a new execution model.<h2>Looking Ahead</h2>These standards are early. AARTS v0.1 is a draft. Skill ID signing is an evolving proposal. Adapter conformance testing, enterprise fail-open profiles, and cross-marketplace reputation sharing remain active areas of collaboration.But the direction is clear.AI agents are moving from experimentation to infrastructure. Security must do the same.The Gen Agent Trust Hub is our contribution to building a predictable, interoperable trust layer for the agentic ecosystem.&nbsp;We invite host builders, marketplace operators, enterprise security teams, and fellow security vendors to review, adopt, and challenge.Because in the agentic era, safety cannot be bespoke. It must be foundational.&nbsp;

Shaping the standards for a safer, freer Agentic Era 

Leading the Way for AI Agent Safety 

Director of Threat Research & Applied AI

We built a security engine that protects AI agents across several major platforms. The hardest part wasn't building the protection. It was discovering that each platform exposes different security-relevant events, provides different context about what the agent is doing and why, and enforces security decisions differently. The same agent action we can catch and block in one platform is invisible in another, not because our detection failed, but because the platform never told us it happened. There are now over a dozen agent platforms, from IDE assistants like Claude Code, Cursor, and GitHub Copilot, to agentic frameworks like LangChain, CrewAI, and OpenAI Agents SDK, to orchestrators like n8n, and each one handles this differently, if at all.Today, Gen is releasing AI Agent Runtime Safety Standard (AARTS) as part of its Agent Trust Hub initiative, which aimsto address this fragmentation. This is not a security product, but rather a set of vendor-neutral standards that connects agent hosts (IDEs, orchestrators, frameworks, standalone agent apps), security engines (components that evaluate agent actions), and adapters (thin mappings that translate host-native events into a common schema). For those building or operating agents, AARTS enables portable, consistent security decisions across hosts, eliminating the need to rebuild security integrations from scratch.&nbsp;The standard is open and vendor-neutral, and any host or security vendor can implement or contribute to it.<h2 class="text-align-justify">The Problem AARTS Addresses</h2>Agentic systems fail in ways that are familiar but occur more rapidly and with greater reach. Examples include sequences evolving from helpful actions to dangerous activities like&nbsp;"download and execute," prompt injection steering tool use, supply chain issues affecting skills and plugins, sensitive data exposure by agents, and permission confusion resulting from sub-agents and external tool providers creating additional trust boundaries. Although these risks are not new, they now manifest within a rapidly diversifying set of hosts and tool APIs.Without a shared contract, security becomes a collection of one-off integrations, each missing context and offering inconsistent enforcement. This leads either to weak coverage, due to incomplete integrations and context, or to overly blunt blocking when trust is undermined by limited interpretation. AARTS targets the root cause: the lack of shared semantics and hook points.In our previous&nbsp;<a href="https://www.gendigital.com/blog/news/company-news/ai-agent-detection-and-response" target="_blank">post on Agent Detection and Response</a>, we described how attackers register AI-hallucinated package names on npm and PyPI, preloaded with infostealers. Catching that requires seeing the package install action with enough context — the package name, the source, whether the agent hallucinated it or the user requested it. One platform gives us all of that. Another gives us the command string with no structured metadata. A third doesn't emit the event at all. Same attack, three different levels of visibility.Simply put, AARTS standardizes where security decisions can be made (hook points), what data is available for decision-making (data model), and how decisions are enforced (verdict semantics), ensuring consistent protections across different agent hosts.<h2 class="text-align-justify">Architecture and Roles</h2>AARTS defines three distinct roles:<ul type="disc"><li data-list-item-id="ef43f78807d68ed5d7cafc41e43ed6a77">Host: The environment that runs the agent. The host emits lifecycle events at defined hook points and enforces decisions.</li><li data-list-item-id="ea8ecfedda9151e1e4ff45954fc0bd003">Adapter: A minimal translation layer that maps host-specific events into AARTS events, normalizing tool names, lifecycle phases, and context.</li><li data-list-item-id="ec84e412023baa1359b60a4c189c0ce73">Security Engine: The evaluator that receives AARTS events, applies detection strategies, and returns structured verdicts.</li></ul>This separation matters. Embedding security logic in hosts results in superficial security ("checkbox security"), while engines depending on host internals compromise portability. AARTS keeps the boundary clean: hosts provide the right moments and context, and engines make decisions. AARTS also supports various integration models (in-process, subprocess, remote), accommodating different latency, isolation, and operational requirements.<h2 class="text-align-justify">How the Standard Works</h2>AARTS defines hook points, moments in an agent's lifecycle where a security engine can intercept, evaluate, and act. Version 0.1 specifies 19 of them across session management, prompt lifecycle, tool execution, plugin and skill loading, sub-agent behavior, memory operations, and output delivery.Not all hooks carry equal weight. Three illustrate the design logic:<ul type="disc"><li data-list-item-id="e9e499b9374a4cd1f71378518f359c0d6">PreToolUse is the primary checkpoint. Before an agent runs a shell command, writes a file, makes a network request, or installs a package, the security engine sees the action, its arguments, and the context that led to it, then returns a verdict: allow, deny, or ask the user. This is where most real-world threats get caught.</li><li data-list-item-id="ebd37c3ab71aca6ef9cddd1080d793f75">PreLLMRequest operates earlier in the loop, before the prompt reaches the model. It lets engines inspect prompt integrity: whether safety instructions survived context compaction, whether untrusted content was injected into the conversation, whether the instruction layering is intact. If PreToolUse is the checkpoint at the door, PreLLMRequest watches for problems before the agent even decides what to do.</li><li data-list-item-id="ecce291342cb701e4290928d827faf930">PreSkillLoad addresses supply chain risk. When an agent installs a plugin or skill from a marketplace, the engine can evaluate the trust source (where it came from, whether it's signed, whether it was modified) before it gets loaded. AARTS can also direct the host to strip hooks from skills, preventing plugins from disabling the security layer that's supposed to govern them.</li></ul>The remaining 16 hooks follow the same pattern: they sit at the moment where risk becomes real and provide the engine with enough context to make a decision.For those decisions to be portable across hosts, the inputs have to be consistent. AARTS standardizes the event structure: every event carries a mandatory envelope (hook point, session ID, timestamp, host identity) plus standard extensions, reusable field groups for conversation context, tool details, trust source metadata, modification tracking, and provenance. A canonical tool taxonomy maps host-specific tool names (Cursor's "terminal" vs. Claude Code's "bash") to common categories (shell, file_write, web_request, package_install), so detection logic written once works everywhere.Verdicts are structured: a decision (allow, deny, or ask), a threat category, a severity level, confidence, and a human-readable explanation for the user. Two design choices are worth noting. If the host can't prompt the user, ask falls back to deny, safe but a real constraint for headless orchestrators. If the engine crashes or times out, the host treats it as allow and logs the failure. Fail-open is the right default for developer tools where latency matters, but enterprise deployments will want to revisit this per environment.<h2 class="text-align-justify">Conformance Levels: Gradual Adoption</h2>AARTS defines three conformance levels:<ul type="disc"><li data-list-item-id="e18c7418214ef4f97b683f20402deb227">Level 1 (Basic): Requires SessionStart and PreToolUse hooks, minimum tool coverage (shell, file_write, web_request), enforcement of allow and deny (ask may be treated as deny).</li><li data-list-item-id="e0369d9837c299bdfd5a5c490fb899318">Level 2 (Standard): Adds plugin and skill load hooks, prompt lifecycle protection, broader tool coverage, additional hooks (PrePluginLoad, PreSkillLoad, PostInputProcessing, PreLLMRequest, PostToolUse), and requires provenance on PreToolUse.</li><li data-list-item-id="e31f70f735daa583aec2ef8c2d180686d">Level 3 (Comprehensive): Provides full lifecycle coverage including user input, prompt response cycle, tool modification, sub-agents, MCP connect, memory read/write, output delivery, and session end. Clarifies sub-agent inheritance and memory protection.</li></ul>This tiered approach offers a realistic adoption path: Level 1 addresses basic failures, Level 2 tackles prompt integrity and supply chain issues, and Level 3 supports persistent and autonomous agents.<h2 class="text-align-justify">Open&nbsp;Considerations</h2>As we move AARTS beyond a draft specification,&nbsp;the following are being considered:<ul type="disc"><li data-list-item-id="ef56e3529222a5671c75f8e80fbc5c2c9">Enterprise fail-open story: The spec mandates fail-open behavior. Deployment profiles may be needed to reflect the realities of different environments.</li><li data-list-item-id="e5e2beee0ceee16ba315173c83dcc4596">Adapter integrity: Ensuring adapters correctly map tool types and extract artifacts is vital. A conformance test suite for adapters is as important as the specification itself is a next-step priority.</li><li data-list-item-id="eea2912c8ee60ab64a8339e4d3c2b5748">Consent for privacy: Explicit consent is required for external transmission of prompts and tool data. Operational guidance or recommended patterns are needed to avoid incompatible implementations.</li><li data-list-item-id="e7dbd49139e599330dbad43c59209bc82">Ask semantics in headless environments: Treating ask as deny by default is safe but may discourage engines from using ask. A standard way for deployment policy to handle ask in these scenarios is needed.</li></ul>AARTS aims to translate&nbsp;these considerations into predictable operational behavior.<h2 class="text-align-justify">Call for Feedback</h2>AARTS v0.1 is a draft intended for early review by host builders, security vendors, and enterprise operators.&nbsp;The full specification is available <a href="https://github.com/gendigitalinc/aarts" target="_blank">here</a>. If you build agent hosts, consider whether you want every security integration to remain one-off. If you develop security engines, ask yourself whether you want to rewrite the same logic for each host and still miss critical context. AARTS aims to make the answer “no”.&nbsp;We welcome contributions — open an issue or submit a pull request.

AARTS standardizes how AI agent platforms expose security events, enabling portable runtime protection across the growing agent ecosystem

Introducing AARTS: An Open Standard for AI Agent Runtime Safety

You gave an AI agent access to your computer. Maybe it's writing code for you, managing your emails, or handling your finances. It's working with your credentials, acting on your behalf, and making decisions you used to make yourself. Fast, confidently, and largely unsupervised. That's the deal: you trade safety and oversight for productivity. That changes the security model.But who's watching what it actually does?That is why we built the Gen Agent Trust Hub (ATH), our trust layer for the agentic AI ecosystem, designed to secure agents across their lifecycle from skill to runtime to enforcement.Agent Detection and Response (ADR) is the runtime pillar — the layer that governs what agents actually do.<h2 class="text-align-justify">What is ADR?</h2>ADR intercepts agent actions at the moment of execution, before they land, and evaluates whether they're safe. It allows, asks the user, or denies. That's the core.Verification tells you what an agent is. ADR tells you what it's doing. These are complementary. Verification is necessary, but it can't govern what happens at runtime. A verified skill can still hallucinate a malicious package name. A signed plugin can still be exploited through prompt injection. A trusted agent can still lose its safety instructions mid-session and go off the rails. The moment of action is where risk becomes real.The principle isn't new. Every major computing paradigm has developed its own detection and response layer. Endpoints got EDR. Distributed infrastructure got XDR. Not replacements for what came before, but purpose-built layers for new execution surfaces with new threat models. AI agents are the next such surface, and ADR is the layer built for it.Within the Agent Trust Hub, ADR is the enforcement engine that turns trust signals into real-world protection.<h2>Why Now?</h2>The threats targeting AI agents are already here. What follows is a small sample of the agentic threats we’re seeing. The actual attack surface spans everything an agent can do, and agents can do a lot.Your agent installs malware for you. AI agents hallucinate. When they hallucinate a package name during installation, that's not just an error, it's an opportunity. Attackers monitor AI-suggested package names and register them on npm and PyPI, pre-loaded with infostealers and backdoors. The technique is called slopsquatting, and it's already being exploited at scale. In February 2026, researchers uncovered a campaign of&nbsp;<a href="https://www.securityweek.com/new-sandworm_mode-supply-chain-attack-hits-npm/">19 typosquatted packages specifically</a> targeting users of AI coding tools, designed to exfiltrate SSH keys, AWS credentials, and API tokens.Your agent ignores your instructions. Summer Yue, Meta's Director of Alignment, told her OpenClaw agent to&nbsp;<a href="https://techcrunch.com/2026/02/23/a-meta-ai-security-researcher-said-an-openclaw-agent-ran-amok-on-her-inbox/">confirm before taking any action</a> on her email. It worked, until the session ran long enough to trigger context compaction, a mechanism where the agent compresses its memory to keep running. The safety instructions were among the first things dropped. The agent deleted 847 emails. She had to physically run to her machine to stop it.Your agent is a new delivery channel for malware. Attackers are already using multiple paths to get malicious code running through agents. Weaponized skills on public marketplaces have delivered the&nbsp;<a href="https://x.com/GenThreatLabs/status/2018323584834318419">AMOS infostealer to macOS users</a>. Malicious project configurations have achieved&nbsp;<a href="https://research.checkpoint.com/2026/rce-and-api-token-exfiltration-through-claude-code-project-files-cve-2025-59536/">remote code execution in Claude Code</a> the moment a developer opens a repository. Prompt injection through fetched content can turn a trusted agent into an attacker's proxy. The delivery mechanisms vary. The result is the same: your agent executes something it shouldn't, and it does so with your permissions.These are just three examples. The pattern is consistent: agents operate with real authority, and that authority is being exploited. Through supply chains, through the agent's own fragility, and through the same malware ecosystem that has targeted every other computing platform before them.<h2>Where ADR Must Operate</h2>To be effective, ADR must sit as close to the agent's execution loop as possible. Not as an external service watching from the outside. Inside, where the actions happen.Agent sessions contain source code, credentials, business logic, and personal data. ADR evaluates locally, on the device. No prompts, tool inputs, or code leave your machine for analysis.Agents are not robust systems. Their own internal mechanisms can silently drop safety instructions or get poisoned across sessions. A security layer that only sees inputs and outputs misses the moment things go wrong inside. ADR watches the execution loop itself: every action the agent attempts, in context, in sequence.Only a client-side layer sees the complete picture: the prompt history, what files were already modified, the sequence of prior actions. A remote service sees isolated events. ADR sees the session. That difference is what makes it possible to distinguish a legitimate file deletion from a compromised agent cleaning up its tracks.<h2>Sage: ADR in Practice</h2>In February, we&nbsp;<a href="https://www.gendigital.com/blog/news/company-news/sage-ai-agent-security">open-sourced Sage</a> as the first implementation of ADR. It runs client-side, directly inside the agent's execution loop, across Claude Code, Cursor, and OpenClaw. Over 200 detection rules cover supply chain attacks, credential exposure, dangerous commands, persistence mechanisms, and more, backed by Gen's threat intelligence.Since launch,&nbsp;<a href="https://github.com/avast/sage/">Sage</a> has quickly crossed 1,000 installs and is available across the major agentic tool marketplaces. The open-source community is already contributing, porting Sage to new platforms like OpenCode and hardening detection rules. As one independent security researcher concluded after&nbsp;<a href="https://www.linkedin.com/pulse/paranoid-sage-testing-security-tool-agents-radek-hendrych-capff/">testing Sage</a>: "Real-time monitoring and management of AI agents based on deterministic measures is, in my view, the right way to go."All these responses confirm what we suspected: people want this layer.<h2>What's Next</h2>ADR is the runtime layer. But runtime protection alone isn't enough.&nbsp;The Agent Trust Hub is built on the idea that trust in the agentic era must be layered: verification before installation, ecosystem trust signals across marketplaces, and enforcement at execution. ADR closes the gap at the moment risk becomes real.&nbsp;The industry needs a common interface for how agents expose their actions to security tools. A shared standard, so that protection works consistently regardless of which agent you use or which platform you're on. We've been working on that too. More soon.

Inside the Gen Agent Trust Hub, ADR turns trust signals into real-world protection.

Gen Agent Trust Hub Introduces Agent Detection and Response: The Missing Security Layer for Agentic AI 

Threat Research Team

AI agents are starting to do the parts of computing that used to be “hands on keyboard”. They run terminal commands, edit files, install packages, and fetch things from the web, because that’s what it takes to get real work done. Tools like Claude Code, Cursor, and OpenClaw give these agents direct access to your machine and your sensitive information. That's what makes them useful. It's also what makes them risky. These agents make mistakes. They install packages that don't exist because they hallucinated the name. They pick up credentials and pass them where they shouldn't. They'll download and run programs from the internet without a second thought if the task seems to call for it. All of this happens fast, with confidence, and without asking. If you’ve watched an agent confidently suggest curl … | bash or “just add your API key here” into a file it shouldn’t touch, you’ve seen the problem.&nbsp;Agents are also targets. Malicious URLs, poisoned dependencies, and compromised plugins can turn an agent into an attack vector, one that already has access to your files, your terminal, and your network. This is already happening. <a href="https://x.com/GenThreatLabs/status/2018323584834318419" target="_blank" rel="noreferrer noopener">We recently identified</a> around 400 malicious skills on ClawHub, a public registry for agent extensions. At the time of discovery, they accounted for 12% of all available skills. These weren’t obviously shady throwaways; many were weaponized versions of legitimate skills, designed to trick agents into downloading malware.&nbsp;&nbsp;These are the same agents we're starting to trust with payments, infrastructure, and sensitive workflows. The gap between what agents can do and what keeps them in check is real, and it's growing.&nbsp;Today, Gen, the company behind Norton, Avast, LifeLock, and AVG, is releasing Sage to close that gap. With Sage, the Agent Trust Hub now evaluates and governs agent behavior during execution. Not just before installation. Not just at verification. But in the exact moment an action is about to happen.&nbsp;

<h2>What Sage does</h2>Think of Sage as a lightweight safety shield for AI agents. It sits inside the agent's workflow and checks every action before it happens: shell commands, URL fetches, file writes, package installs. For each action, Sage returns a verdict:&nbsp;<ul type="disc"><li data-list-item-id="e2aee30f293d4edf72390d310cdfa3f0e">Allow — no threat detected, the action proceeds&nbsp;</li><li data-list-item-id="e89d49ae9fae752dee97f089112c466b8">Ask — something looks suspicious, the user decides&nbsp;</li><li data-list-item-id="e44ca3f76876a9e666f2544e1233366ba">Deny — confirmed threat, the action is blocked&nbsp;</li></ul>The agent never notices unless something is wrong.&nbsp;Sage catches both sides of the risk. It protects agents from external threats like malicious URLs, compromised packages, and dangerous downloads. And it protects users from agents themselves, when they try to run destructive commands, expose credentials, or write to sensitive system files.&nbsp;Here's a concrete example. An agent automating a setup task decides to download and execute a script from the internet. No review, no confirmation. Sage recognizes the pattern and blocks it before anything runs.&nbsp;<h2 class="text-align-justify">Part of a larger commitment</h2>Sage is not a standalone effort. It’s part of a broader strategy to embed continuous trust into the AI skills ecosystem through the Gen Agent Trust Hub.&nbsp;&nbsp;The Skill Scanner of the <a href="https://ai.gendigital.com/agent-trust-hub" target="_blank" rel="noreferrer noopener">Gen Agent Trust Hub</a> helps determine whether an agent extension is trustworthy before it gets installed, using cloud-based verification and classification. Sage picks up where that leaves off. Once the agent is running, it enforces safety locally, on your machine, checking every action at the moment it's about to execute.&nbsp;&nbsp;Together, they cover the full lifecycle: from the decision to install something to the moment it actually runs. More capabilities are on the way as the agentic ecosystem grows.&nbsp;<h2 class="text-align-justify">Open source, on purpose</h2>The agentic platforms themselves are designed for open integration: open-source tools, community plugins, public hook APIs. A closed security layer would be harder to adopt in a space that moves fast and depends on community integrations.&nbsp;At the same time, there is no established playbook for how to secure AI agents. By open-sourcing Sage, we're proposing one and inviting the community to help define what safe agents should look like in practice.&nbsp;This first version is lightweight. It's not a full antivirus product as we know it, but it addresses the problems where they are. It covers the high-risk paths we think matter most: commands, file changes, URL fetches, and package installs, including catching hallucinated or malicious dependencies before they land on your machine. Sage ships with over 200 detection rules covering command injection, persistence, credential exposure, obfuscation, and supply-chain attacks.&nbsp;Sage currently supports Claude Code, Cursor, and OpenClaw, with more platforms on the way. If you find a detection that's too noisy or not strict enough, open an issue. If you want to add a rule, improve a check, or add support for another agent platform, send a pull request.&nbsp;We built Sage to be extended.&nbsp;&nbsp;<h2>Try it</h2>Learn more about Sage at <a href="https://ai.gendigital.com/sage" target="_blank" rel="noreferrer noopener">https://ai.gendigital.com/sage</a>. For technical details, documentation, and source code, visit the project on <a href="https://github.com/avast/sage" target="_blank" rel="noreferrer noopener">https://github.com/avast/sage.</a>&nbsp;If you use Claude Code, Cursor, or OpenClaw, install Sage and put it in the path of real work. The fastest way to improve a safety layer is to expose it to real commands, real repos, and real mistakes. That's what we're doing here — shipping early, in the open.&nbsp;

Closing the growing security gap between what Al agents can do and what keeps them in check

Introducing Sage: Safety for Agents 

AI agents aren’t just helpful in chats anymore. They’re actually taking action on your behalf. Autonomous AI can access data. They move from suggestion to execution without hesitation.&nbsp;This is what we call “mindless artificial intelligence.”&nbsp;When technology starts taking action on our behalf, trust has to be built into the system itself.&nbsp;That’s why Gen is partnering with Vercel to bring independent safety verification to the AI skills ecosystem at <a href="https://skills.sh/" target="_blank">skills.sh</a>, embedding trust directly into the skills layer.&nbsp;

<h2>Independent verification. Built in.&nbsp;</h2>Through this partnership, the Gen <a href="https://ai.gendigital.com/agent-trust-hub" target="_blank">Agent Trust Hub</a> will evaluate skills on skills.sh and assign clear, transparent classifications:&nbsp;🟢 Safe&nbsp; 🟡 Low Risk&nbsp; 🟠 High Risk&nbsp; 🔴 Critical Risk&nbsp;Behind those ratings is advanced risk modeling and threat intelligence from Gen Threat Labs, analyzing permissions, behaviors, vulnerabilities and potential malicious intent.&nbsp;The goal isn’t to slow innovation. It’s to help it scale responsibly. Developers should be able to build boldly. Users should be able to install confidently. And both deserve clarity.&nbsp;<h2>Security as infrastructure, not an afterthought&nbsp;</h2>By integrating verification directly into skills.sh, Gen and Vercel are making security part of the AI development lifecycle, not something bolted on after the fact.&nbsp;In the agentic world, skills are how things get done. Until now, developers and users had limited visibility into whether a skill was truly safe. Reputation and documentation can’t tell the whole story. As agents gain autonomy, guesswork isn’t good enough.&nbsp;The trust has to be visible.

<h2>Powering Digital Freedom in what comes next&nbsp;</h2>At Gen, we believe Digital Freedom means being able to embrace new technology without sacrificing safety or confidence. As AI evolves, our responsibility evolves with it.&nbsp;The future of AI will be powerful. And expansive. But, it must also be secure by design.&nbsp;This partnership is one more step toward building the trust foundation the agentic era demands, so innovation and confidence can grow together.&nbsp;

Bringing transparent risk ratings to the AI skills ecosystem to help developers build boldly and users move confidently

The trust layer for the agentic era 

Today, we’re proud to announce that we’ve joined the&nbsp;<a href="http://globalantiscam.org/">Global</a><a href="https://www.gasa.org/"> Anti-Scam Alliance </a>(GASA) as a Foundation Member — a critical next step in our Purpose of Powering Digital Freedom.<h2>Why GASA — and why Gen?</h2>The Global Anti-Scam Alliance (GASA) exists to build a coordinated, international front against digital scams. It brings together governments, consumer protection organizations, financial institutions, tech platforms and cybersecurity leaders to do what no single player can do alone: share intelligence, shape policy, and deliver rapid, systemic action.Gen brings to the table:<ul type="disc"><li>Global reach across 150+ countries and nearly 500 million users</li><li>A multi-brand portfolio including Norton, LifeLock, Avast, AVG, Avira, CCleaner, and ReputationDefender</li><li>Cutting-edge innovation, like&nbsp;<a href="https://genie.norton.com/">Norton Genie</a>, our AI-powered scam detection tool that helps people spot malicious content in real-time<ul><li>A longstanding commitment to education, privacy, and consumer empowerment — backed by ongoing scam research and insights from our team.&nbsp;<a href="https://www.gendigital.com/blog/insights">Read our insights blog</a>.&nbsp;</li></ul></li></ul>By joining GASA, we’re extending that impact — working with fellow members to build smarter, faster, more connected defenses that match the sophistication of today’s scam operations.<h2>This is about more than technology. It’s about trust.</h2>Online scams are no longer isolated incidents. They are systemic, sophisticated, and increasingly powered by emerging technologies like AI. They mimic real conversations, exploit trusted brands, and adapt faster than many institutions can respond.&nbsp;We’ve tracked this shift in our own&nbsp;<a href="https://www.gendigital.com/blog/insights/reports/threat-report-q4-2024">research and reporting</a>.&nbsp;There’s a stigma around being scammed. And it keeps people quiet. We’re here to change that — by shifting the conversation and strengthening the support systems that help people take back control.Because falling for a scam isn’t a sign of weakness. It’s a sign of just how advanced and manipulative scam tactics have become. And if we want to reverse the trend, we need to meet it with collaboration, transparency and relentless innovation.<h2>What’s next?</h2>Joining GASA is not just a partnership. It’s a signal of intent.We intend to work across borders, sectors and systems to drive meaningful change. We intend to build tools that don’t just react but anticipate. And, we intend to empower people with the clarity, context and confidence they need to protect themselves in a complex digital world.We’re honored to join this alliance&nbsp;as we deepen our commitment to protecting people. We’re advocating for smarter, more proactive digital safety policies worldwide. Visit our&nbsp;<a href="https://www.gendigital.com/us/en/public-policy/">Public Policy</a> page to learn more.&nbsp;

Leyla Bilge

Posts by Leyla Bilge

Love in the age of algorithms

Joining the global anti-scam alliance