Sydney researchers uncover a hidden limit to AI predictability
What happened
Researchers at the University of Sydney, led by Professor Pulin Gong and PhD student Andrew Ly, discovered a fundamental mathematical reason why AI systems can exhibit different behaviours even when trained under almost identical conditions. Their study, published in Nature Machine Intelligence, shows that tiny differences in a model’s starting conditions can lead to divergent outcomes despite deterministic training environments.
Why it matters
This finding highlights a hidden limit to AI predictability that is rooted in the mathematics of neural network learning, suggesting that improving AI reliability may require addressing these intrinsic instabilities.
Is this a big deal?
Get stories like this every morning
The day's AI news in plain English, in your inbox at 8:30 IST. Free.
More AI news
Every Way to Get 192GB for Local AI (Without a $9,500 Mac) [video]Industry news · 14 Nov 2026DeepSeek V4 Pro 0813DeepSeek · 14 Nov 2026DeepSeek V4 Pro 0813DeepSeek · 14 Nov 2026Show HN: Portal2Bench – Can AI agents solve Portal 2?Industry news · 14 Nov 2026Anthropic Upgrades Claude Code Projects to Run Parallel AI Workers Hands-FreeAnthropic · 14 Nov 2026This page was collected from sydney.edu.au and summarised automatically by AIPulsee's pipeline. See how it runs.