Gary Marcus

Rank 19 of 47
|
Score 64
6/24/2025, 5:56:18 PM
X
In reply to:
Geoffrey Miller
@primalpoly
·
12d
@GaryMarcus @S_OhEigeartaigh Are you suggesting, Gary, that EAs concerned about AI safety should try to _accelerate_ AI capabilities development by finding better alternatives to RL-based LLMs, so we can develop AGIs faster?

Speaking for myself, I'd much rather see AI companies waste as much time and money
Gary Marcus
@GaryMarcus
·
14d
@S_OhEigeartaigh sure, but the money has been there to support alternatives to LLMs. but that’s not how things went.
Seán Ó hÉigeartaigh
@S_OhEigeartaigh
·
14d
@GaryMarcus I hope you're sitting for this as it might come as a shock, but not all EAs approve of Sam Bankman-Fried's actions and decision-making.
Gary Marcus
@GaryMarcus
·
14d
@S_OhEigeartaigh FLI alone has close to a billion.

EA SBF put 500M intro anthropic.

zero effort that i can see towards anything but directly and indirectly accelerating LLMs.
Seán Ó hÉigeartaigh
@S_OhEigeartaigh
·
14d
@GaryMarcus The companies are far-better funded than AI safety community, & are putting billions of dollars into advancing capabilities in this paradigm - makes sense for AI safety community to try to address the problems. Makes less sense for them on the margin to try to come up w a whole
Gary Marcus
@GaryMarcus
·
14d
For reasons unknown to me, the AI Safety community has put almost all of its eggs into scaling + system prompts + RL.

Judging by how many problems we are seeing now (see below) with models built per that formula, shouldn’t we be desperately trying to find alternatives?

The statement engages in a discussion about the potential risks and consequences of giving more power and agency to large language models (LLMs), which is a significant public issue related to AI safety and ethics. The tone is cautionary, highlighting concerns about the potential harm and catastrophic consequences of current AI development practices.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement raises concerns about potential harm from LLMs, aligning with the principle of striving to do no harm by advocating for caution. [+2]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement aims to promote understanding of the risks associated with LLMs, fostering a dialogue about AI safety. [+2]
  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 addressing the topic of AI safety without resorting to personal attacks. [+1]
  4. Principle 6:
    I will use my influence for the betterment of society.
    By discussing the potential societal impacts of LLMs, the statement uses influence to raise awareness about AI safety issues. [+2]
  5. Principle 7:
    I will uphold the principles of free speech and use my platform responsibly and with integrity.
    The statement contributes to the discourse on AI safety, using the platform to responsibly discuss the implications of AI development. [+2]