People have always looked for patterns to explain the universe and to predict the future. “Red sky at night, sailor’s delight. Red sky in morning, sailor’s warning” is an adage predicting the weather.

It is an increasingly familiar experience. A request for help to a large language model (LLM) such as OpenAI’s ChatGPT is promptly met by a response that is confident, coherent and just plain wrong.

AI is ubiquitous now—from interpreting medical results to driving cars, not to mention answering every question under the sun as we search for information online. But how do we know it is safe to use, ...