AI security veteran maps the debate: security walls agents in, alignment keeps them honest
joshua_saxe · x · 2026-09-04
Joshua Saxe, founder of an AI security company, attempts to synthesize the fragmented AI security conversation, which he describes as feeling like a "tower of Babel":
- Security's job: put the agent in a padded room, monitor everything it does, and ensure it neither tries to escape nor misuses its freedoms. Mostly achievable with current tools, but gaps remain around non-human identities, agent-specific monitoring tech and practices, and sandboxing that can contain agents with superhuman hacking abilities.
- Alignment's job: ensure the agent never even tries to do anything illegal or immoral — though since alignment can never offer deterministic guarantees, it remains fundamentally a probabilistic bet.
The post is a framework-level attempt to bring order to a debate where different communities talk past each other.
Related event: Security veterans untangle AI safety: security vs alignment vs policy(9 posts)→
More from AGI Musings
- AI agents are erasing work-life boundaries: managing agents on commutes, vacations and at 3am — bendee983 · 2026-10-21
- a16z: Superhuman AI hardware design is the easy part — manufacturing is the hard one — a16z · 2026-10-21
- a16z's Casado says AI can't do a VC's job — but it can do everything around it — alvelda · 2026-10-21
- Kent Beck: Let the Genie Carry the Toil, But Keep the Struggle — josh_wills · 2026-10-21
- Nearly $3B mapped: the money behind the AI existential risk ecosystem — KeyNaive2982 · 2026-10-21
- Three reasons AI is not conscious and may never be, per Nature essay — anilkseth · 2026-10-21