Sorry really can be the hardest word.

It’s hard for ordinary people to distinguish good advice from decisively-voiced bad advice.

A new technique automatically guides an LLM toward outputs that adhere to the rules of whatever programming language or other format is being used.

Large language model AIs can ingest long documents and answer questions about them, but a key question is how well they ‘understand’ individual sentences in the documents.

A new method from the MIT-IBM Watson AI Lab helps large language models to steer their own responses toward safer, more ethical, value-aligned outputs.

Aligning AIs with people’s goals and values is tricky. A new technique quantifies how far off human and machine are from each other.

Even experts get dog body language wrong at times.

Even experts get dog body language wrong at times.

When AI systems try to bridge gaps in their training data, the results can be wildly off the mark: fabrications and non sequiturs researchers call hallucinations.

Handing over the tasks once done by human developers comes with some major risks.