91av

Technology

Changing a single number among billions can destroy an AI model

Today's huge AI models are composed of several billion numbers known as weights and changing just one of them can destroy their ability to function, leading to “gibberish” output

By Matthew Sparkes

26 November 2024

Abstract illustration of artificial intelligence

The output of AIs is controlled by the interactions of huge networks of nodes or weights

shulz/Getty Images

An artificial intelligence model can be made to spout gibberish if a single one of the many billions of numbers that compose it is altered.

Large language models (LLMs) like the one behind OpenAI’s ChatGPT contain billions of parameters or weights, which are the numerical values used to represent each “neuron” of their neural network. These are what get tuned and tweaked during training so the AI can learn abilities such as generating text. Input is passed through these weights, which determine the most statistically likely output.…

Sign up to our weekly newsletter

Receive a weekly dose of discovery in your inbox! We'll also keep you up to date with 91av events and special offers.

Sign up

To continue reading, today with our introductory offers

or

Existing subscribers

Sign in to your account