Can AI accelerate AI itself? Let's look at what actually happened!!!
AI autonomously discovered a new RL algorithm, invented 100+ novel architectures, and evolved a data curation strategy β each one outperforming the best human-designed counterparts:
π€ RL algorithm design β discovered novel optimization mechanisms that beat human-designed GRPO by +12.5 pts on AMC
π§ Neural architecture search β invented 100+ new architectures, surpassing Mamba, GatedDeltaNet, and more by +0.97 pts over SOTA
π¦ Pretraining data curation β evolved a pipeline that outperforms FineWeb-Edu, DCLM, and Nemotron-CC by +18 pts on MMLU
𧬠Drug-target interaction β discovered a new architecture that outperforms all prior human-designed methods by +6.94 AUROC
These aren't toy demos. These are frontier-level results β produced without a human in the loop.
This is ASI-Evolve: an open-source agentic framework that closes the loop between knowledge β hypothesis β experiment β analysis, and repeats it autonomously until it finds something that works. We built it for AI research. But the loop doesn't care about the domain.
A year ago, we made a prediction:
Scale computation β Scale scientific discovery.
The world caught up faster than we expected.
Which raises a question we can't stop thinking about β and we think you shouldn't either:
In the post-AGI era, when AI generates breakthroughs faster than humanity can absorb them β when the rate of discovery permanently outpaces the rate of value conversion β what becomes the new scarcity?
Not intelligence. Not data. Not compute.
Something far more irreplaceable.
We don't have the full answer. But we think it's worth asking β loudly, and together.
π» Code: https://t.co/rBicTvkxYT
π Paper: https://t.co/2DriXJ4uFl (1/8)
β Star it. Fork it. Point it at your hardest problem.
π¨ We can download models, but not see how they were built.
Introducing daVinci-LLM: most transparent LLM pretraining project.
π Open source: model weights, data pipeline, training process, ablations.
π― 3B model matching 7B performance
π Report: https://t.co/HgTcXhSQSS (1/7)