OpenAI says its new internal AI model has resolved more than 100 long-standing mathematical problems, surprising its own mathematicians and raising questions about whether humans can keep up.
Mozilla is highlighting a new framework for understanding open source AI, arguing that model weights, training data, code, documentation, and other components can have very different levels of openness.
OpenAI’s new LifeSciBench benchmark was supposed to measure how useful AI can be in scientific research. Instead, it also highlights just how far today’s most advanced models still have to go before they can be trusted with real science.