Gary Marcus

Rank 15 of 47
|
Score 75
12/19/2024, 2:19:57 PM
X
In reply to:
Jacob Aron
@jjaron
·
277d
@GaryMarcus The researchers did actually check this (worth noting that it's not an OpenAI paper either) https://www.newscientist.com/art…
Gary Marcus
@GaryMarcus
·
277d
You have to read very far into the replies in thread to discover that the cases may be in the training set.

Caveat emptor.
Deedy
@deedydas
·
279d
o1-preview is far superior to doctors on reasoning tasks and it's not even close, according to OpenAI's latest paper.

AI does ~80% vs ~30% on the 143 hard NEJM CPC diagnoses.

It's dangerous now to trust your doctor and NOT consult an AI model.

Here are some actual tasks:

1/5

The statement is part of a discussion on the reliability and accuracy of AI in medical diagnostics, which is a significant public issue. The conversation involves evaluating AI's capabilities compared to human doctors, raising questions about trust and safety in healthcare.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not directly cause harm but is part of a broader discussion that could influence perceptions of AI in healthcare.
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The conversation promotes understanding by discussing the complexities of AI in diagnostics, though it could benefit from more balanced perspectives. [+1]
  3. Principle 4:
    I will engage in constructive criticism and dialogue with those in disagreement and will not engage in personal attacks or ad hominem arguments.
    The dialogue is constructive, focusing on evidence and data rather than personal attacks. [+1]