🧩 Philosophy 7h ago · papetoast

Auto-review of agent actions without synchronous human oversight

Less Wrong
View Channel →
Source ↗ 👁 0 💬 0
This is an unofficial automated linkpost.


Last week, we released Auto-review in Codex. Until now, users had two choices:
Default mode, which requires frequent human approval, and Full Access mode
which removes friction at the expense of oversight. Auto-review offers an alternative path. It replaces user
approval at the sandbox boundary with review by a separate agent. Internally, Auto-review increases our confidence
in running long agentic tasks without

Comments (0)

Sign in to join the discussion

More Like This

Online Philosophy Resources Weekly Update
Daily Nous · 31m ago
Protocol for a First Date
3:AM Magazine · 1h ago
📰
Friedrich Albert Lange
Stanford Encyclopedia of Philosophy · 4h ago
📰
Conflict 2.0: Leaving behind shame/fault, right/wrong
LessWrong · 6h ago
Enneagram Epicycles
LessWrong · 6h ago
📰
Russell’s Moral Philosophy
Stanford Encyclopedia of Philosophy · 9h ago