OpenAI briefly halved GPT-6 Astra's hallucination rate in post-launch benchmark edit
The company also boosted its own math scores and cut Anthropic's, raising questions about 'benchmaxxing' in AI evaluation.
By Turki Al-Mutairi·· 3 min

Loading the Newsroom
Curating from trusted global sources…
5 briefings · “benchmark”
The company also boosted its own math scores and cut Anthropic's, raising questions about 'benchmaxxing' in AI evaluation.
The Eternal Sunshine Tour wrapped at London's O2 with fewer dates but far bigger nightly yields, resetting the benchmark for a blockbuster live run.
Big IPO, bigger options debut: what it means for investors, risk teams, and anyone benchmarking market appetite.
SpaceX shares jumped, and Musk’s $800B pre-IPO value crossed a trillion, reshaping how investors price “moonshots.”
The horror hit crossed a massive box office line faster than most arthouse films ever do, reshaping what A24 can do in theaters.