Geoffrey Irving Welcomes Public AI Team Disagreements

Geoffrey Irving, Chief Scientist at the UK AI Security Institute, posted a reply expressing satisfaction that a post reveals internal disagreements among AI teams. He states that teams differ not only on which research may succeed but also on whether specific work counts as positive at all, citing capability externalities. Irving has held prior roles at DeepMind, OpenAI, and Google Brain focused on alignment. The reply appears as a screenshot attached to the visible post on X.