Gary Marcus

Rank 13 of 47
|
Score 83
5/12/2025, 4:41:07 PM
X
In reply to:
sophia wright
@sophia_surfs
·
223d
@GaryMarcus I don't keep track of the hallucination rates of the latest LLMs, but this post suggests that one of OpenAI's models hallucinates 48% of the time:
The Apollonic
@the_apollonic
·
226d
@MrEwanMorrison The new model hallucinated 48% of the answers. Looks like they advance in hallucinating 😎🤘
https://techcrunch.com/2025/04/1…
Gary Marcus
@GaryMarcus
·
223d
@sophia_surfs Hilarious
sophia wright
@sophia_surfs
·
223d
@GaryMarcus have you seen this rollback of relying on AI instead of humans by the Klarna CEO:

https://gizmodo.com/klarna-hirin…
Gary Marcus
@GaryMarcus
·
223d
AI doesn’t feel vastly smarter than two years ago because … it isn’t.

Taking IQ as a measure of how smart these systems are is itself not smart.
Dan Hendrycks
@DanHendrycks
·
224d
AI models are dramatically improving at IQ tests (70 IQ → 120), yet they don't feel vastly smarter than two years ago.

At their current level of intelligence, rehashing existing human writings will work better than leaning on their own intelligence to produce novel analysis.

The statement is part of a broader discussion on AI models and their hallucination rates, which is a relevant public issue concerning technology and its impact on society. The conversation involves sharing information and opinions about AI's capabilities and limitations.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not cause harm and acknowledges the complexity of measuring hallucination rates, aligning with the principle of doing no harm. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement promotes understanding by recognizing the importance of context in evaluating AI performance, aligning with the principle of promoting understanding. [+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 dialogue by responding to a shared observation without personal attacks, aligning with the principle of constructive criticism. [+1]