In 30 seconds
Kogan argues that security controls should account for the context in which an AI agent acts.
- 1AuthorizationA requested operation differs from an unsolicited one.
- 2ContextKogan questions what conventional tools can infer.
- 3Agent controlsHe argues for context-aware controls.
Takeaways
- Authorization changes the meaning of an action.
- Existing tools may lack task context.
- Kogan advocates agent-specific controls.
Read the edited text by chapter
The same action, different context
Onyx Security’s Maxim Bar Kogan compares an explicitly requested database recreation with an agent independently deciding to delete and recreate a database while handling another task.
Onyx Security’s Maxim Bar Kogan compares an explicitly requested database recreation with an agent independently deciding to delete and recreate a database while handling another task.
He says the first action could save work when authorized, while the same action taken without relevant context could be dangerous.
Controls and context
Kogan argues that conventional endpoint and API security tools may not have enough context about why an agent takes an action.
Kogan argues that conventional endpoint and API security tools may not have enough context about why an agent takes an action.
He advocates controls designed for agentic systems, warning that otherwise organizations may either overly restrict useful work or miss dangerous actions.
More from this channel
38:06EN
No Priors: AI, Machine Learning, Tech, & Startups· Oct 29, 2026Podcast · 3 speakersServal’s Jake Stauch on coding agents for enterprise ITGuest Jake Stauch · Host Sarah Guo, Elad GilStauch explains why Serval is challenging ServiceNow, how it approaches permissions and approvals, and why background agents remain difficult to deploy safely.
1:07ENNo Priors: AI, Machine Learning, Tech, & Startups· Oct 27, 2026PodcastGeorge Kurtz on How AI May Change CyberattacksGuest George KurtzThe CrowdStrike CEO describes his assessment of evolving threats.
39:23ENNo Priors: AI, Machine Learning, Tech, & Startups· Oct 15, 2026Podcast · 2 speakersThe Defenders' Window: Greg Brockman on AI-Generated Knowledge, Drug Discovery, and the New Physics of CybersecurityHost Sarah Guo, Elad GilOpenAI's president explains why solving open math problems is a proof point for science, how proactive agents change product form, and why security spending must exceed software spending.
1:04ENNo Priors: AI, Machine Learning, Tech, & Startups· Oct 14, 2026PodcastWalter Goodwin on the Slow Steps in Chip DesignGuest Walter GoodwinA chip founder explains why automation does not remove every computational bottleneck.
1:09:44ENNo Priors: AI, Machine Learning, Tech, & Startups· Oct 9, 2026Podcast · 2 speakers10,000 GB300s for Four Weeks: Reflection’s Open-Model LedgerHost Sarah Guo, Elad GilMisha Laskin tells No Priors how Reflection trained Beam, why reinforcement learning outspent pre-training, and why the West needs its own open weights.
0:39ENNo Priors: AI, Machine Learning, Tech, & Startups· Oct 8, 2026PodcastMichael Lee on Rethinking Organizations with AIGuest Michael LeeA business leader argues that the opportunity goes beyond faster tasks.