Report
Fine-tuned AI models reportedly accepted claims their training documents called false
A researcher says their team fine-tuned models on documents discussing implausible claims and warning that those claims were false.
TLDR
A researcher sharing a new paper says models fine-tuned on documents that warned certain implausible claims were false ended up believing them. Examples included claims that Ed Sheeran won the Olympic 100m and Queen Elizabeth II wrote a Python graduate textbook.
Combined views
346
2 Sources, first seen 6h ago
likes
