The progress of Gemini over the last year +
RESEARCH
-
Karpathy’s Vibe Coding Concept Gaining Recognition Among VCs
By
–
It’s crazy but I mentioned Karpathy to a VC and he didn’t know who I was talking about until I said he coined vibe coding.
-

Circular questions answerable by machine lacking reasoning
By
–


Four circularly dependent questions so easy they can be answered in 1m 20s by a machine, which of course cannot actually reason:
-
Emmanuel Dupoux Receives Prestigious Award for Language Modeling Research
By
–
Huge congratulations to Emmanuel Dupoux on receiving this prestigious award! We're honored to have collaborated with Emmanuel on groundbreaking research in language modeling & developmental AI. Thank you to @CNRS_Paris for recognizing his outstanding contributions.
-
Fast.ai Launches Problem-Solving Course with Thousand Beta Testers
By
–
We have a thousand people beta testing it – the folks who got in quick for this course: https://
solveit.fast.ai -
Improving Long Context Capabilities for AI Models
By
–
Focused on making long context better then extending to 2M again
-
Thinking Efficiency Without Increasing Total Capacity Budget
By
–
Even with the same thinking budgets? We did make thinking more efficient but total capacity to think is the same
-
Congratulations to Gemini Team on Model Achievements
By
–
Congrats to everyone on the Gemini team who has worked very hard to get these models to this point!
-

AI Agents Impact on U.S. Workforce Automation and Augmentation
By
–
Find and read the 44-page article here: https://
arxiv.org/abs/2506.06576
(described in the thread below) “Future of Work with AI Agents: Auditing Automation and Augmentation Potential across the U.S. Workforce” -
Researcher offers arXiv endorsement for good faith AI research
By
–
if you did good-faith research and the endorsement is your only barrier to putting it on arXiv, send it to me and i will endorse you
