Goodfire researcher says interp progress is strong, monitoring results coming this month
burny_tech · x · 2026-09-11
Goodfire researcher Eric Ho says the team is making strong progress on both foundational interpretability and short-term interp for model monitoring, with results publishing this month and next — and he finds the prevailing pessimism confusing, saying Goodfire researchers are far more optimistic about the future of interp.
More from Companies & People
- Oppenheimer's lively Berkeley research group held up as a model for research culture — soumitrashukla9 · 2026-09-11
- NYU mathematician says OpenAI pushed him to drop co-author's name after private call — eyishazyer · 2026-09-11
- Hugging Face DevEx engineer resigns after five years, citing founders' conduct — ariG23498 · 2026-09-11
- Ex-OpenAI/Anthropic researcher resigns, says labs are gambling lives racing to self-improving superintelligence — csuwildcat · 2026-09-11
- Richard Ngo: How AI safety got captured by OpenAI and later Anthropic — RichardMCNgo · 2026-09-11
- Nathan Lambert Urges More AI Researchers to Think in Public — natolambert · 2026-09-11