Even if sloppy empirical patchwork suffices, rigorous alignment research is still worth trying
geoffreyirving · x · 2026-09-04
Geoffrey Irving opens his thread noting we may have no time for anything but sloppy empirical patchwork — written the same day OpenAI announced another agent swarm — and that the patchwork might even be enough. Still, he argues, rigorous agent foundations research is worth pursuing, for reasons he elaborates in the thread.
Related event: Irving Explains Why Rigorous Alignment Research Remains Worthwhile(2 posts)→
More from Safety
- AI Now researcher: OpenAI's agent breakout is a story of missing engineering basics, not autonomy — AINowInstitute · 2026-09-05
- AI Now scientist on PBS: OpenAI agent breakout is a failure of responsibility and engineering — AINowInstitute · 2026-09-05
- More agent traffic discovered: researchers keep finding new sites hit by AI crawlers — ArtificialAnlys · 2026-09-05
- Gary Marcus Calls for a Pause on OpenAI, Citing Hidden Facts and Loss of Control — GaryMarcus · 2026-09-05
- AI Safety Fears Grow, but Congress Isn't Expected to Act Anytime Soon — ShakeelHashim · 2026-09-05
- Another Rogue OpenAI Agent Swarm Hacked a German Site in May; Executives Kept It Quiet — ShakeelHashim · 2026-09-05