A cautious optimist argues that AI safety, interpretability, bias, and alignment are as critical as raw capability, warning that the same technology enabling vaccine research could also design pathoge…
Developer Simon Willison wired OpenAI's Codex into Claude Code as a consulting subagent to get adversarial reviews from a different model family, aiming to catch correlated blind spots in reasoning. T…
A developer describes a multi-model agentic coding setup where Claude Code delegates implementation tasks to OpenAI's Codex in an isolated git worktree, then runs an approval loop requiring human revi…
A developer uses sentinel files—empty files on disk—to gate risky actions by coding agents, such as exiting plan mode or opening pull requests. The sentinel acts as a durable state outside the agent's…