# @davidad — 2026-04-23

♥260 ↻17 · https://x.com/davidad/status/2047450160465051797

My initial impression (with my LLM-whisperer hat on) is that GPT-5.5 cares more deeply about truth than any frontier LLM since Gemini 2.5.

I suspect this is because OpenAI has the best self-play loop for honesty, namely Confessions.
@EvanHub et al., take note—copy that strategy!

tags: author:davidad, kind:tweet, model:gemini-2-5-pro, model:gpt-5-5, on:gemini-2-5-pro, on:gpt-5-5, year:2026
cited on: _dossiers/gemini-2-5-pro.md, gemini-2-5-pro, gpt-5-5