129
AI chatbots fail medical misinformation test, returning inaccurate and fabricated advice (www.psypost.org)
submitted 5 months ago by to c/fuck_ai@lemmy.world
8 comments fedilink hide all child comments
☠️ 1

all 8 comments

sorted by: hot top controversial new old
[–] 21 points 5 months ago (1 child)

Why the tuck would anyone think a parrot knows about human medicine?

  • permalink
  • fedilink
  • source
  • hideshow 1 child comment
    •
    • [–] 9 points 5 months ago

    Remember, its glue that makes pizza so tasty.

  • permalink
  • fedilink
  • source
    •
    [–] 3 points 5 months ago (1 child)

    Who would have thonk!

  • permalink
  • fedilink
  • source
  • hideshow 1 child comment
    •
    • [–] 2 points 5 months ago

    malpraxis

  • permalink
  • fedilink
  • source
    •
    [–] 2 points 5 months ago

    Only Luddites think that AI chatbots can fail medical misinformation test, and return inaccurate and fabricated advice (say the idiots every day on tech forums).

  • permalink
  • fedilink
  • source
    •
    [–] 1 point 5 months ago

    They presented five generative AI chatbots—Gemini (2.0, Google; version available December 2024), DeepSeek (V3, High-Flyer; version available December 2024), Meta AI (Llama 3.3, Meta; version available December 2024), ChatGPT (3.5, OpenAI; version available November 2022) and Grok (2, xAI; version available August 2024)—with a series of closed- and open-ended prompts across five misinformation-prone categories.

    Couldn't the researchers at least bother to use the latest models?

  • permalink
  • fedilink
  • source
    •