Researchers Discover GPT-4 Is Less Accurate Than GPT-3

  • Researchers have found that GPT-4, OpenAI’s latest version of its large language model (LLM), is less accurate than GPT-3.
  • According to analysis from NewsGuard, GPT-4 is worse than its predecessor in terms of spewing lies as truths.
  • The researchers published their findings in a report.


Sign Up for nextbigwhat newsletter

The smartest newsletter, partly written by AI.

Download, the short news app for busy professionals