Claims of GPT-6 Astra scoring 98.6% on ARC-AGI-3 don’t hold up to scrutiny
Covered by 1 source · 1 article
Unverified claims about GPT-6 Astra's performance highlight the challenges of assessing AI advancements and the need for transparent benchmarks. The post Claims of GPT-6 Astra scoring 98.6% on ARC-AGI-3 don’t hold up to scrutiny appeared fi…
Covered by
