OpenAI’s o3 system scored 85% on the ARC-AGI benchmark, well above the previous AI best score of 55% and on par with the average human score.

Over the past 25 years, technological innovation has accelerated unprecedentedly, transforming societies worldwide.

The company says the algorithm can tackle problems across fields such as programming, physics and math. It’s based on another ...

A new set of much more challenging evals has emerged in response, created by companies, nonprofits, and governments. Yet even ...

Until models like ChatGPT can learn from small numbers of examples and adapt with more sample efficiency, they will only be ...