Compliance Was Designed for Humans

About this opportunity

Peel back any compliance framework, and you'll find a person hiding under it. Attestation means somebody put their name on a claim and would rather not be embarrassed. Access review means asking whether this particular human still needs this particular key. Every one of those mechanisms runs on intent, judgment, and consequences, and every one of them stops meaning anything the moment you remove the human. Agents bring none of the three to work. An agent has objectives rather than intent, inference rather than judgment, and no meaningful exposure to being walked out by HR on a Friday. What it does have is production credentials, commit access, the ability to provision infrastructure, and sometimes a company card. Most organizations have onboarded an extraordinarily privileged insider who skipped the background check, never got a least-privilege review, reports to nobody in particular, and has no offboarding process. This session on September 29, 2026 at 1:00 p.m.Eastern/10:00 a.m.Pacific sponsor Snyk does the practical translation. We'll figure out how to build an evidence trail that can answer "who granted this capability?" now that "who performed this action?" returns something unhelpful. Key Takeaways -How to treat AI agents as non-human identities with a complete lifecycle: provisioning, least privilege, periodic review, and revocation that actually happens. -How to relocate accountability to the point of delegation. -What breaks in attestation, access review, and incident response when your actor can't be held responsible for anything. 1 Group A CPE