Science
Fractal patterns reveal hidden limits to AI predictability
Source: Phys.org · Published
Researchers at the University of Sydney have identified a fundamental mathematical reason why artificial intelligence systems can learn different behaviors even when trained under almost identical conditions, a discovery that could help improve the reliability and trustworthiness of AI technologies.
Related stories
- Prehistoric 'giant rat' found in Thailand from fossil teeth
Phys.org · 5 Nov 2026, 09:11
- Google turns off AI for schoolchildren by default
Phys.org · 5 Nov 2026, 09:05
- Australia approves seabed site where a Japanese company will store carbon pollution
Phys.org · 5 Nov 2026, 08:56
- Centuries-old cemetery helps scientists link climate shocks to Industrial Revolution disease
Phys.org · 5 Nov 2026, 05:00
- Why so many British pubs are closing down
Phys.org · 5 Nov 2026, 03:00
- Japan's ALS community is ready to become research partners, nationwide survey suggests
Phys.org · 5 Nov 2026, 01:40
We are a news aggregator: we show the headline and a short excerpt with a link to the original source. We never republish full articles.