B, a 3-billion-parameter AI model, is challenging OpenAI, Google and DeepSeek on math and coding benchmarks while reigniting the debate over AI scaling, benchmark gaming and small-model reasoning.

What if your biggest challenges could become your greatest teachers? Learn five ACT-based ways to work with problems instead ...

Phil Chen, a fromer researcher at OpenAI and how the founder of an agent-native startup, has offered fresh career guidance ...

The bees had to roll the ball under a blue "flower," then stand atop the moved object to access a sweet treat. Mikko Törmänen / University of Oulu Some bumblebees can spontaneously solve problems, a ...

Many students appear to be completing assignments faster while learning less from them. This conclusion comes from one of the largest studies of how generative AI is changing student behavior and ...

At Data + AI Summit, Databricks CEO Ali Ghodsi unveiled LTAP, a new architecture that collapses the 40-year unification ...

Bigger has defined AI from day one. New data says task-specific small models beat frontier LLMs on accuracy, cost and speed — ...

These short anomaly-detection puzzles are designed to illustrate how reasoning often depends on identifying inconsistencies ...

The second batch of “First Proof” problems is meant to evaluate AI’s usefulness for research-level math. The best model got six or seven of the ten questions right.

Think about placing dots on a flat surface. You want as many pairs as possible to be separated by the same distance. For any amount of dots, what is the greatest possible number of pairs that can be ...

DeepMind’s AlphaProof system solved four out of six problems at the 2024 International Mathematical Olympiad, generating formally verified proofs through reinforcement learning trained on millions of ...