Google SLAMS OpenAI's GPT-5: This Is EMBARASSING!
OpenAI researchers claimed groundbreaking AI achievements in solving Erdos problems, but Google DeepMind's leader dismissed it as embarrassing, suggesting the solutions were merely advanced Google searches.
MAIN POINTS FROM TRANSCRIPT
- OpenAI claimed GPT5 solved 10 Erdos problems and made progress on 11 more.
- Mark Sulk announced finding errors in Erdos's original work, sparking excitement.
- Google DeepMind's leader criticized the claims as embarrassing.
- Thomas Bloom, Erdos problems curator, provided a reality check on the claims.
TAKEAWAYS
- OpenAI's announcement was used as a recruitment tool.
- The AI community is divided on the validity of OpenAI's claims.
- Google DeepMind's reaction highlights competitive tensions in AI advancements.
- The incident underscores the importance of verifying AI research claims.