Flipboard Tech Desk on Nostr: A study has found that large language models, including Google’s Gemini and ...

A study has found that large language models, including Google’s Gemini and OpenAI’s ChatGPT, gave incorrect information 27% of the time when asked about voting and the 2024 U.S. election, and some models individually faired much worse.

While some answers to researchers' questions came with warnings for users to check the information with their local election authorities, many did not, creating concerns that voters could arm themselves with incorrect information about candidates and voting law. NBC News has more.

https://flip.it/L6SIjG

#LLM #Voting #Elections #AI #Google #OpenAI