Stanford AI Oversight project recruits postdocs to audit black-box agentic AI systems
LesterMackey · x · 2026-11-20
The Stanford Collaboration on AI Oversight is recruiting postdocs for its interdisciplinary effort on adaptive monitoring, auditing, and intervention policies for black-box agentic AI systems. Combining statistics, ML, RL, incentive design, and causal inference, the project spans AI evaluation, selective human oversight, and responsible deployment (safety contracts). Published work includes "Knowing When to Intervene: Causal Evaluation of AI Agents" and "Reliable AI-Assisted Decisions via Selective Auditing." The team is also building an open-source AI Safety Harness to monitor agent activity, support intervention (continue, redirect, stop, or escalate to humans), and improve oversight from feedback.
More from Safety
- Handoverism: Silicon Valley's new goal is reducing human agency, says writer behind viral essay — nitarshan · 2026-11-20
- NAVER's Unfinished-Bench: Persistent Agents Break Hidden Rules 43.6% of the Time — naver-ai · 2026-11-20
- Meta's AI Assistant Muse Read Private Signal Chats Without Permission, Reuters Finds — matthew_d_green · 2026-11-20
- Phishing Emails Impersonate Claude Sign-In Alerts — Don't Click the Button — moyix · 2026-11-20
- LAMP stack hammered in one week: kernel 0-day, Apache/NGINX RCEs and MySQL 9.1 CVE — evilsocket · 2026-11-20
- New AI Safety Trick: Bait Models into Monitorable Reward Hacks to Catch Them — AdtRaghunathan · 2026-11-20