This AI ‘mechanic’ can peer inside the black box of other neural networks

MIT researchers have developed an AI model capable of interpreting and editing the internal workings of other neural networks. This 'mechanic' model can predict the function of specific components and improve the performance of other AI systems.
Why it matters
This breakthrough offers a potential solution to the 'black box' problem in AI, allowing for greater transparency and control over how complex models make decisions.
Search author on: PubMed Google Scholar
Bluesky Facebook LinkedIn Reddit Whatsapp X Save article View saved research Some AI models behave like mechanics — fixing issues in other models that they neither designed nor fully understand. Credit: Mint Images/Getty
Researchers have used an artificial-intelligence model to read and tinker with the internal workings of another AI network — offering a new way to peek into the ‘black box’ of AI .
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in