Gary Marcus

Rank 16 of 47
|
Score 66
12/19/2024, 3:30:26 PM
X
In reply to:
Jacob Aron
@jjaron
·
276d
@GaryMarcus No, but our reporter spoke to them about it
Gary Marcus
@GaryMarcus
·
276d
@jjaron it’s in a supplement. do you have the actual supplement?
Jacob Aron
@jjaron
·
276d
@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
·
276d
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
·
278d
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 and conversation involve a discussion about the reliability and transparency of AI in medical diagnostics, which is a significant public issue. The tone is critical, highlighting concerns about the accessibility and visibility of important information regarding AI training data.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement raises concerns about transparency, which can help prevent harm by ensuring accurate information is available. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    By questioning the visibility of critical information, the statement encourages a deeper understanding of AI's role in healthcare. [+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 statement engages in constructive criticism by pointing out a potential issue in the presentation of data. [+1]