


Researchers are figuring out how large language models work
Such insights could help make them safer, more truthful and easier to use
TO MOST PEOPLE, the inner workings of a car engine or a computer are a mystery. It might as well be a black box: never mind what goes on inside, as long as it works. Besides, the people who design and build such complex systems know how they work in great detail, and can diagnose and fix them when they go wrong. But that is not the case for large language models (LLMs), such as GPT-4, Claude and Gemini, which are at the forefront of the boom in artificial intelligence (AI).
LLMs are built using a technique called deep learning, in which a network of billions of neurons, simulated in software and modelled on the structure of the human brain, is exposed to trillions of examples of something to discover inherent patterns. Trained on text strings, LLMs can hold conversations, generate text in a variety of styles, write software code, translate between languages and more besides.
Models are essentially grown, rather than designed, says Josh Batson, a researcher at Anthropic, an AI startup. Because LLMs are not explicitly programmed, nobody is entirely sure why they have such extraordinary abilities. Nor do they know why LLMs sometimes misbehave, or give wrong or made-up answers, known as “hallucinations”. LLMs really are black boxes. This is worrying, given that they and other deep-learning systems are starting to be used for all kinds of things, from offering customer support to preparing document summaries to writing software code.
Explore more
This article appeared in the Science & technology section of the print edition under the headline “Inside the mind of an AI”

A scientific discovery could lead to leak-free period products
Polymers from algae can turn menstrual blood into a gel

Vaccines could keep salmon safe from sea lice
A successful jab would be a boon to fish farmers

New yeast strains can produce untapped flavours of lager
One Chilean hybrid has a spicy taste, with hints of clove

A scientific discovery could lead to leak-free period products
Polymers from algae can turn menstrual blood into a gel

Vaccines could keep salmon safe from sea lice
A successful jab would be a boon to fish farmers

New yeast strains can produce untapped flavours of lager
One Chilean hybrid has a spicy taste, with hints of clove
A new technique could analyse tumours mid-surgery
It would be fast enough to guide the hands of neurosurgeons
The world’s most studied rainforest is still yielding new insights
Even after a century of research, Barro Colorado in Panama continues to shed light on natural life
A new bionic leg can be controlled by the brain alone
Those using the prosthetic can walk as fast as those with intact lower limbs