--- title: The Lethal Trifecta Is Already on Your Laptops — Michael... description: A conference talk from a Coder solutions engineer explains the 'lethal trifecta' concept coined by Simon Willison: AI agents become dangerous when they combine... canonical: https://daily.dev/posts/the-lethal-trifecta-is-already-on-your-laptops-michael-patterson-coder-brk8w7dld twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: The Lethal Trifecta Is Already on Your Laptops — Michael Patterson, Coder | daily.dev og:description: A conference talk from a Coder solutions engineer explains the 'lethal trifecta' concept coined by Simon Willison: AI agents become dangerous when they combine... og:url: https://daily.dev/posts/the-lethal-trifecta-is-already-on-your-laptops-michael-patterson-coder-brk8w7dld og:image: https://api.daily.dev/og/posts/BrK8w7DLd.png og:image:alt: The Lethal Trifecta Is Already on Your Laptops — Michael Patterson, Coder og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # The Lethal Trifecta Is Already on Your Laptops — Michael Patterson, Coder **[AI Engineer](https://daily.dev/sources/aidotengineer)** · 20 min read · 0 upvotes · 0 comments ## Summary A conference talk from a Coder solutions engineer explains the 'lethal trifecta' concept coined by Simon Willison: AI agents become dangerous when they combine access to private data, exposure to untrusted content, and the ability to communicate externally. The talk covers attack scenarios like prompt injection, context poisoning, and privilege escalation, then pitches Coder's remote development environments, model proxy servers, and agent-level firewalls as mitigations. It also discusses organizational challenges like 'shadow AI' where employees use unsanctioned AI tools, and what CISOs and DevOps teams worry about when scaling agentic AI across a company. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the lethal trifecta in AI agent security? The lethal trifecta, a term coined by Simon Willison, describes three combined risk factors that make AI agents dangerous: access to private data, exposure to untrusted content such as support tickets or logs, and the ability to communicate externally to the internet. Any one factor alone is manageable, but having two is risky and all three together is considered unacceptable in a secure corporate environment, since an attacker can inject instructions that cause the agent to exfiltrate data. _Teams rolling out coding agents can follow ongoing coverage of agent security practices on daily.dev._ ### How can I prevent an AI coding agent from deleting my production database or leaking files? Run agents in an isolated remote or virtual environment rather than directly on a laptop, limit external network communications so the agent can only reach approved domains, and restrict credentials and permissions to only what the task requires. Adding a model proxy to log and scrub traffic to the LLM provider, plus a process-level firewall that blocks unapproved commands like destructive database or repo deletions, further contains agent mistakes or prompt injection attacks. _Developers wiring up agent workflows can track hardening techniques like these on daily.dev._ ### What is shadow AI and why is it a security risk for companies? Shadow AI is the unauthorized use of AI tools and personal subscriptions by employees when a company has not officially approved agentic AI, similar to shadow IT with unsanctioned devices or software. It is risky because agent-based AI run outside company oversight can cause damage quickly, and at scale this becomes 'ninja AI' - many unmonitored agents taking actions across an organization until something breaks. _Security leads weighing AI governance policies can keep up with these risks via daily.dev._ ## Similar posts on daily.dev - [The Lethal Trifecta: Why Your AI Agent Could Be a Security Liability](https://daily.dev/posts/the-lethal-trifecta-why-your-ai-agent-could-be-a-security-liability-4nfnrjhax) · Snowflake Community · 1 upvotes · 0 comments - [Agentic AI and Security](https://daily.dev/posts/agentic-ai-and-security-zqrngwfxj) · Martin Fowler · 40 upvotes · 3 comments - [Agents are rewriting the rules of security. Here’s what engineering needs to know.](https://daily.dev/posts/agents-are-rewriting-the-rules-of-security-here-s-what-engineering-needs-to-know--khtu9taim) · The New Stack · 0 upvotes · 0 comments - [Fragments: September 24](https://daily.dev/posts/fragments-september-24-nzgkchulg) · Martin Fowler · 0 upvotes · 0 comments --- Tags: [#ai-security](https://daily.dev/tags/ai-security), [#prompt-injection](https://daily.dev/tags/prompt-injection) [View this post on daily.dev](https://daily.dev/posts/the-lethal-trifecta-is-already-on-your-laptops-michael-patterson-coder-brk8w7dld) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"The Lethal Trifecta Is Already on Your Laptops — Michael Patterson, Coder","url":"https://daily.dev/posts/the-lethal-trifecta-is-already-on-your-laptops-michael-patterson-coder-brk8w7dld","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/the-lethal-trifecta-is-already-on-your-laptops-michael-patterson-coder-brk8w7dld"},"datePublished":"2026-10-10T15:13:46.886Z","dateModified":"2026-10-15T02:25:18.176Z","description":"A conference talk from a Coder solutions engineer explains the 'lethal trifecta' concept coined by Simon Willison: AI agents become dangerous when they combine...","image":"https://i.ytimg.com/vi/abw_m-DRi1k/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/abw_m-DRi1k/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/the-lethal-trifecta-is-already-on-your-laptops-michael-patterson-coder-brk8w7dld","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-security,prompt-injection","timeRequired":"PT20M","video":{"@type":"VideoObject","name":"The Lethal Trifecta Is Already on Your Laptops — Michael Patterson, Coder","description":"A conference talk from a Coder solutions engineer explains the 'lethal trifecta' concept coined by Simon Willison: AI agents become dangerous when they combine...","thumbnailUrl":"https://i.ytimg.com/vi/abw_m-DRi1k/sddefault.jpg","uploadDate":"2026-10-10T15:13:46.886Z","duration":"PT20M","url":"https://api.daily.dev/r/BrK8w7DLd","embedUrl":"https://www.youtube.com/embed/abw_m-DRi1k"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"The Lethal Trifecta Is Already on Your Laptops — Michael Patterson, Coder"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/the-lethal-trifecta-is-already-on-your-laptops-michael-patterson-coder-brk8w7dld#faq","mainEntity":[{"@type":"Question","name":"What is the lethal trifecta in AI agent security?","acceptedAnswer":{"@type":"Answer","text":"The lethal trifecta, a term coined by Simon Willison, describes three combined risk factors that make AI agents dangerous: access to private data, exposure to untrusted content such as support tickets or logs, and the ability to communicate externally to the internet. Any one factor alone is manageable, but having two is risky and all three together is considered unacceptable in a secure corporate environment, since an attacker can inject instructions that cause the agent to exfiltrate data. Teams rolling out coding agents can follow ongoing coverage of agent security practices on daily.dev."}},{"@type":"Question","name":"How can I prevent an AI coding agent from deleting my production database or leaking files?","acceptedAnswer":{"@type":"Answer","text":"Run agents in an isolated remote or virtual environment rather than directly on a laptop, limit external network communications so the agent can only reach approved domains, and restrict credentials and permissions to only what the task requires. Adding a model proxy to log and scrub traffic to the LLM provider, plus a process-level firewall that blocks unapproved commands like destructive database or repo deletions, further contains agent mistakes or prompt injection attacks. Developers wiring up agent workflows can track hardening techniques like these on daily.dev."}},{"@type":"Question","name":"What is shadow AI and why is it a security risk for companies?","acceptedAnswer":{"@type":"Answer","text":"Shadow AI is the unauthorized use of AI tools and personal subscriptions by employees when a company has not officially approved agentic AI, similar to shadow IT with unsanctioned devices or software. It is risky because agent-based AI run outside company oversight can cause damage quickly, and at scale this becomes 'ninja AI' - many unmonitored agents taking actions across an organization until something breaks. Security leads weighing AI governance policies can keep up with these risks via daily.dev."}}]} ```