Read the CCP's AI+ plan here: https://
gov.cn/zhengce/conten
t/202508/content_7037861.htm?mc_cid=31b20242ff
…
RESEARCH
-
China’s CCP AI+ Plan Released and Available for Review
By
–
-
Google’s AI Decline Amid Cultural Politics Debate
By
–
“Google gone too woke and hence lost AI” is such a funny timeline in hindsight As if the only thing that matters is the constant cultural warfare that gives US political parties power
-
Hinton’s Safety Measures May Paradoxically Increase AI Risks
By
–
Ironically the stuff Hinton pushes for is what makes this more likely. 🙁
-
Dario Amodei’s Growing Hand-Waving on AI Claims
By
–
The amount of handwaving by Dario Amodei is growing exponentially. pic.twitter.com/Q0qYbj7IF2
— Pedro Domingos (@pmddomingos) 7 septembre 2025The amount of handwaving by Dario Amodei is growing exponentially.
-
LLM Hallucinations Persist Due to Fundamental Fact Understanding Gap
By
–
Hallucinations persist because LLMs don’t know what a fact is. No amount of tweaking the evaluation will change that.
-

Language Models Hallucinate Due to Their Architecture
By
–
No, language models hallucinate because they’re language models.
-

GPT-5 Pro Advances Novel Mathematics with Academic Partnership
By
–
gpt-5 pro for novel mathematics — in partnership with a math professor:
-
Anthropic’s Undisclosed Use of Brave Search Partnership
By
–
Anthropic use Brave but they were very resistant to openly stating that, I had to dig though their sub-processors to confirm it
-
OpenAI and Anthropic Secrecy About Search Index Sources
By
–
It's wild to me how secretive OpenAI and Anthropic are about which search index they're using – it's such important information to help evaluate how trustworthy their results are