@d_m_ziegler Profile picture

Daniel Ziegler

@d_m_ziegler

Alignment Stress-Testing @ Anthropic

Joined April 2011
Similar User
Rachel Freedman photo

@FreedmanRach

Ajeya Cotra photo

@ajeya_cotra

Evan Hubinger photo

@EvanHub

Collin Burns photo

@CollinBurns4

Joe Carlsmith photo

@jkcarlsmith

Alex Turner photo

@Turn_Trout

Ryan Kidd photo

@ryan_kidd44

Cas (Stephen Casper) photo

@StephenLCasper

Daniel Filan @ NeurIPS research-tweets photo

@dfrsrchtwts

Kamal Ndousse photo

@kandouss

Catherine Olsson photo

@catherineols

Elizabeth Barnes photo

@BethMayBarnes

Adam Gleave photo

@ARGleave

Thomas Woodside 🫜 photo

@Thomas_Woodside

Zac Kenton photo

@ZacKenton1

Daniel Ziegler Reposted

We’re starting a Fellows program to help engineers and researchers transition into doing frontier AI safety research full-time. Beginning in March 2025, we'll provide funding, compute, and research mentorship to 10–15 Fellows with strong coding and technical backgrounds.

AnthropicAI's tweet image. We’re starting a Fellows program to help engineers and researchers transition into doing frontier AI safety research full-time.

Beginning in March 2025, we'll provide funding, compute, and research mentorship to 10–15 Fellows with strong coding and technical backgrounds.

Daniel Ziegler Reposted

Guys, I don't often ask you to retweet, but please retweet this. Swap Your Vote *does not have enough safe state voters to match all its swing state voters!* Swap Your Vote (link below) matches swing state voters who prefer Harris to Trump but don't want to vote for Harris...


Daniel Ziegler Reposted

SwapYourVote is a very interesting idea -- matches two safe-state Dems with one swing state voter considering going third-party, they switch, increasing overall third-party vote share and Harris's odds in swing states.

This post is unavailable.

Daniel Ziegler Reposted

I'd just like to point out that *I* left OpenAI waaay before it was cool


Daniel Ziegler Reposted

Announcing Transluce, a nonprofit research lab building open source, scalable technology for understanding AI systems and steering them in the public interest. Read a letter from the co-founders Jacob Steinhardt and Sarah Schwettmann: transluce.org/introducing-tr…


Daniel Ziegler Reposted
michael_nielsen's tweet image.

Daniel Ziegler Reposted

John von Neumann:

michael_nielsen's tweet image. John von Neumann:

Daniel Ziegler Reposted

Familiar things (that still don't exist): whistleblower protections for AI employees, mandatory 3rd party testing and result-sharing...

GarrisonLovely's tweet image. Familiar things (that still don't exist): whistleblower protections for AI employees, mandatory 3rd party testing and result-sharing...

Daniel Ziegler Reposted

Incredible indeed! A much bigger deal than I would have guessed, and super interesting—thanks very much for the pointer

diviacaroline's tweet image. Incredible indeed! A much bigger deal than I would have guessed, and super interesting—thanks very much for the pointer
diviacaroline's tweet image. Incredible indeed! A much bigger deal than I would have guessed, and super interesting—thanks very much for the pointer

Daniel Ziegler Reposted

A big part of my job these days is to think about what technical work Anthropic needs to do to make things go well with the development of very powerful AI. I digested my thinking on this, plus some of the Anthropic zeitgeist around it, into this piece: sleepinyourhat.github.io/checklist/

sleepinyourhat's tweet image. A big part of my job these days is to think about what technical work Anthropic needs to do to make things go well with the development of very powerful AI.

I digested my thinking on this, plus some of the Anthropic zeitgeist around it, into this piece:
<a style="text-decoration: none;" rel="nofollow" target="_blank" href="https://t.co/dXAwiUNI6I">sleepinyourhat.github.io/checklist/</a>

I wish people would use prediction markets more for practical decision-relevant questions like this one manifold.markets/DanielZiegler/…


Daniel Ziegler Reposted

If Germany had stuck with nuclear power it would have saved €350 billion and cut emissions by 73% more since 2022. tandfonline.com/doi/full/10.10…


Daniel Ziegler Reposted

This is a really accessible, good-faith, and IMO persuasive FAQ from Yoshua Bengio -- here are my favorite excerpts: - the core case for concern - timelines - why companies would build something so unsafe - the compatibility of reducing catastrophic and more immediate harms

trevposts's tweet image. This is a really accessible, good-faith, and IMO persuasive FAQ from Yoshua Bengio -- here are my favorite excerpts:
- the core case for concern
- timelines
- why companies would build something so unsafe
- the compatibility of reducing catastrophic and more immediate harms
trevposts's tweet image. This is a really accessible, good-faith, and IMO persuasive FAQ from Yoshua Bengio -- here are my favorite excerpts:
- the core case for concern
- timelines
- why companies would build something so unsafe
- the compatibility of reducing catastrophic and more immediate harms
trevposts's tweet image. This is a really accessible, good-faith, and IMO persuasive FAQ from Yoshua Bengio -- here are my favorite excerpts:
- the core case for concern
- timelines
- why companies would build something so unsafe
- the compatibility of reducing catastrophic and more immediate harms
trevposts's tweet image. This is a really accessible, good-faith, and IMO persuasive FAQ from Yoshua Bengio -- here are my favorite excerpts:
- the core case for concern
- timelines
- why companies would build something so unsafe
- the compatibility of reducing catastrophic and more immediate harms

Daniel Ziegler Reposted

if we play our cards right we're all going live to the end of time and have the most wonderful adventures before it all fades away. and everything matters as much as anything ever could matter because it all comes down to what we do


Daniel Ziegler Reposted

The CA Assembly Judiciary has taken this seriously. After citing this letter at length, they suggested an amendment to SB 1047 strengthening whistleblower protections. Here's an excerpt...

A group of current, and former, OpenAI employees - some of them anonymous - along with Yoshua Bengio, Geoffrey Hinton, and Stuart Russell have released an open letter this morning entitled 'A Right to Warn about Advanced Artificial Intelligence'. righttowarn.ai

AndrewCurran_'s tweet image. A group of current, and former, OpenAI employees - some of them anonymous - along with Yoshua Bengio, Geoffrey Hinton, and Stuart Russell have released an open letter this morning entitled 'A Right to Warn about Advanced Artificial Intelligence'.
<a style="text-decoration: none;" rel="nofollow" target="_blank" href="https://t.co/uQ3otSQyDA">righttowarn.ai</a>


Daniel Ziegler Reposted

The details matter! Our target is not safety cases in the abstract, but specific details about how safety cases might look for AI control, interp, scalable oversight, guaranteed safety, process supervision, weak-to-strong, automated safety, ELK, alignment evals, <yours>, etc.


Daniel Ziegler Reposted

If you feel interested in jhanas or other serious meditation practice, I think it's worth your time to read this thread and its replies.

I really wish you would caveat this sort of thing when you evangelize this practice, especially later jhanas / anything to do with cessation. 1/ x.com/nickcammarata/…



Daniel Ziegler Reposted

What about "we should be honest about our enormous uncertainty and the wide range of outcomes on the table, and run labs with a strong internal safety culture while working hard internationally to avoid a second Cold War or an arms race"?


Loading...

Something went wrong.


Something went wrong.