AI21, an Israeli startup building its own large language models (LLMs), is raising a $300 million funding round, according to ...

For organizations with clearly defined problems and verifiable answers, RFT offers a compelling way to align models.

In the OpenAI CEO's first congressional hearing in two years, Altman faced few hostile questions or pushback, as lawmakers' ...

New AI systems are less accurate than older versions, with error rates up to 79%. Learn why this happens and how to protect ...

Hallucinations are an intractable problem in AI. Even the most capable models make things up sometimes, a feature of their ...

LLMs that affirm delusions likened to "schizophrenia-seeking missiles." Billionaire claims AI leaders prepping for chance AI ...

One of the more terribly named models, o4-mini drops the “GPT” element of the naming scheme and awkwardly swaps the 4o around ...

Bloomgarden named 2025 Pinnacle Award Honoree at Medical Marketing + Media Pinnacle Awards ...

OpenAI's GPT-o3 and o4-mini models hallucinate fake facts in up to 79% of answers, raising major concerns about AI accuracy ...

But OpenAI's report highlights some harrowing results for anyone who takes ChatGPT responses at face value. OpenAI found that ...

It doesn't reason. It doesn't understand. It just predicts what comes next. That’s what makes large language models powerful ...