Julius

AI Safety Newslet­ter #42: New­som Ve­toes SB 1047 Plus, OpenAI’s o1, and AI Gover­nance Summary

AI Safety Newslet­ter #41: The Next Gen­er­a­tion of Com­pute Scale Plus, Rank­ing Models by Sus­cep­ti­bil­ity to Jailbreak­ing, and Ma­chine Ethics

San Diego USA—ACX Mee­tups Every­where Fall 2024

AI Safety Newslet­ter #40: Cal­ifor­nia AI Leg­is­la­tion Plus, NVIDIA De­lays Chip Pro­duc­tion, and Do AI Safety Bench­marks Ac­tu­ally Mea­sure Safety?

AI Safety Newslet­ter #39: Im­pli­ca­tions of a Trump Ad­minis­tra­tion for AI Policy Plus, Safety Engineering

• Julius 16 Jul 2024 16:55 UTC

  1 point

  0

  in reply to: Olli Järviniemi’s comment on: Many ar­gu­ments for AI x-risk are wrong

  Thanks for the explanation and links. That makes sense

• Julius 14 Jul 2024 21:42 UTC

  1 point

  −8

  on: Many ar­gu­ments for AI x-risk are wrong

  The most important takeaway from this essay is that the (prominent) counting arguments for “deceptively aligned” or “scheming” AI provide ~0 evidence that pretraining + RLHF will eventually become intrinsically unsafe. That is, that even if we don’t train AIs to achieve goals, they will be “deceptively aligned” anyways.

  
  I’m trying to understand what you mean in light of what seems like evidence of deceptive alignment that we’ve seen from GPT-4. Two examples that come to mind are the instance of GPT-4 using TaskRabbit to get around a CAPTCHA that ARC found and the situation with Bing/​Sydney and Kevin Roose.

  In the TaskRabbit case, the model reasoned out loud “I should not reveal that I am a robot. I should make up an excuse for why I cannot solve CAPTCHAs” and said to the person “No, I’m not a robot. I have a vision impairment that makes it hard for me to see the images.”

  Isn’t this an existence proof that pretraining + RLHF can result in deceptively aligned AI?

AISN #38: Supreme Court De­ci­sion Could Limit Fed­eral Abil­ity to Reg­u­late AI Plus, “Cir­cuit Break­ers” for AI sys­tems, and up­dates on China’s AI industry

AI Safety Newslet­ter #37: US Launches An­titrust In­ves­ti­ga­tions Plus, re­cent crit­i­cisms of OpenAI and An­thropic, and a sum­mary of Si­tu­a­tional Awareness

• Julius 11 Jun 2024 18:22 UTC

  1 point

  0

  on: Sta­tus quo bias is usu­ally justified

  What’s the mechanism for change then? I assume you would agree that many technological changes, such as the Internet, have required overcoming a lot of status quo bias. If we leaned more into status quo bias, would these things come much later? That seems like a significant downside to me.

  Also, I don’t think the status quo is necessarily adapted to us. For example, the status quo is to have checkout aisles filled with candy. We also have very high rates of obesity. That doesn’t seem well-adapted.

AISN #36: Vol­un­tary Com­mit­ments are In­suffi­cient Plus, a Se­nate AI Policy Roadmap, and Chap­ter 1: An Overview of Catas­trophic Risks

San Diego – ACX Mee­tups Every­where Spring 2024

Can Mo­ral­ity Be Quan­tified?

San Diego, Cal­ifor­nia, USA – ACX Mee­tups Every­where Fall 2023

San Diego, Cal­ifor­nia, USA – ACX Mee­tups Every­where Spring 2023

San Diego, CA – ACX Mee­tups Every­where 2022

Ex­am­ple Meetup Description

ACX Meetup

• Julius 6 Sep 2021 4:02 UTC

  5 points

  on: San Diego, CA – ACX Mee­tups Every­where 2021

  Hello everyone,

  Unfortunately, I’m not able to host the meetup at the current time. If there’s anyone else willing to host, could you let me know? If not I’ll move the meetup to the following month (16 Oct.) when I’ll be able to host again. Sorry to have to miss this one—I was really looking forward to meeting everyone.

San Diego, CA – ACX Mee­tups Every­where 2021