ai isn’t hitting a wall. see:
RESEARCH
-
Machine Intelligence Doesn’t Need Human-Like Design
By
–
good points – harkens to the view that machine intelligence doesn’t have to be just like human intelligence to be mega significant and useful for us
-

O3 Performance Shows Rapid Progress in RL Chain-of-Thought Scaling
By
–
o3 is very performant. More importantly, progress from o1 to o3 was only three months, which shows how fast progress will be in the new paradigm of RL on chain of thought to scale inference compute. Way faster than pretraining paradigm of new model every 1-2 years
-

OpenAI o3 Breakthrough Reasoning Model Enters Safety Testing
By
–
o3, our latest reasoning model, is a breakthrough, with a step function improvement on our hardest benchmarks. we are starting safety testing & red teaming now.
-

Rapid Progression of AI Models: Toward Superintelligence by 2030
By
–
September 2024: o1-preview had an IQ of 120 December 12, 2024: o1 displayed an IQ of 133 o3, successor to o1, just gained 15 points in mathematics. It becomes increasingly plausible that AI could be a million times smarter than humans by 2030
-
Meta releases Video Seal model with permissive license and code
By
–
With this release we’re making the Seal model available under a permissive license, alongside a research paper, training code and inference code.
-
Meta Introduces Video Seal: Neural Video Watermarking Framework
By
–
Introducing Meta Video Seal: a state-of-the art comprehensive framework for neural video watermarking. Try the demo ➡️ https://t.co/oX7ZlbP5nc pic.twitter.com/7VorrZdnz5
— AI at Meta (@AIatMeta) 20 décembre 2024Introducing Meta Seal: a state-of-the art comprehensive framework for neural video watermarking. Try the demo https://
go.fb.me/1ihqor -
OpenAI Launches o3-mini and o3 Models for Safety Testing
By
–
if you are a safety researcher, please consider applying to help test o3-mini and o3. excited to get these out for general availability soon. extremely proud of all of openai for the work and ingenuity that went into creating these models; they are great.
-
OpenAI Launches o3 and o3 Mini for Safety Testing
By
–
OpenAI announcing o3 and o3 mini today! Not launching, but allowing public access for safety testing. It scores 71% on SWE BENCH—over 20% better than o1 Coding is going to be changed forever
