A naukri.com initiative
Dev
7d
334
Image Credit: Dev
AI Now Passes Turing Test: GPT-4 Fools Human Judges in Landmark Study
- Large language models (LLMs) including GPT-4 are now indistinguishable from humans in blind Turing tests.
- Evaluators correctly identified the source (human vs AI) only 49.9% of the time - no better than random guessing.
- GPT-4 was often judged as more human than actual humans.
- The "pass" of the Turing test represents a significant milestone in AI development.
Read Full Article
20 Likes
For uninterrupted reading, download the app