Academic Summarization: LLMs have been found to fabricate study results, blend findings from unrelated papers or invent ...

While reassembling those pieces isn’t trivial, there is early evidence that LLMs might make it far easier. LLM agents could ...

Using generative AI to design, train, or perform steps within a machine-learning system is risky, argues computer scientist Micheal Lones in a paper appearing in Patterns. Though large language models ...

GBH Morning Edition host Mark Herz spoke with MIT computer science professor Marzyeh Ghassemi about AI's use in medicine.

The growing field of machine unlearning aims to make large language models forget harmful information without retraining them ...

Using generative AI to design, train, or perform steps within a machine-learning system is risky, argues computer scientist Micheal Lones in a paper publishing April 22 in the Cell Press ...

Instead of just predicting words, world models actually learn how the physical world works, which is the "common sense" AI ...

Some fear frontier LLMs like Mythos and Anthropic's GPT-5.5 will lead to cybersecurity annihilation. Ari Herbert-Voss notes ...

In 2016, an AI program he developed at Google DeepMind, AlphaGo, taught itself to play the famously difficult game of Go with ...

Hackers and evildoers are using adversarial poetry to jailbreak AI. The trick involves writing poems as prompts. AI ...

AI researchers refer to this as “model collapse,” a phenomenon in which models trained on their own synthetic outputs degrade ...

The best thing about self-hosted LLMs is that you can choose from hundreds of models ...