Twitxy Saved

Laura Weidinger

@weidingerlaura

temporarily back on X bc these times. @weidingerlaura.bsky.social. Researcher at @deepmind | Measuring and Evaluating AI | AI Ethics | Views my own
321 Following 2.4K Followers 582 Posts

Latest posts 5

Pinned post

Laura Weidinger @weidingerlaura
If you're an AI employee wondering how you could leverage your role for better outcomes than what we're currently seeing -- check out concernedaistaff.org. You won't find all the answers, but a strong, cross-lab community that is working to make a difference <3
concernedaistaff.org Coalition of Concerned AI Staff Our coalition helps AI employees act on their concerns about AI, with the support from colleagues across labs and independent groups.
pamela mishkin Verified account @manlikemishap Sep 10
if you work at an AI company, you might be asking if you, too, should leave. i hope that, if you stay, you're also asking what about how to use your leverage in staying.

since february, a group of us have been working with staff across labs to help answer this question. (1/2)
View quoted post on X
2 replies 6 reposts 12 likes 1.6K views View on X
Laura Weidinger @weidingerlaura
📣 Are LLMs aligned to Human Rights? We’ve just released the first technical evaluation of this question: arxiv.org/abs/2502.19463, led by @rafiyaj111

Check out our results 🧵⬇️

Very excited that this important work is now out -- congratulations to all involved!
arxiv.org Hedging and Non-Affirmation: Quantifying LLM Alignment on... Hedging and non-affirmation are behaviors exhibited by large language models (LLMs) that limit the clear endorsement of specific statements. While these behaviors are desirable in subjective...
Rafiya @rafiyaj111 Feb 28, 2025
(1/7) Excited that our preprint lnkd.in/eRAnDwZm is finally out! 📰Our work demonstrates the first evaluation of LLM alignment to the universal declaration of human rights (UDHR), the most widely internationally recognized doc about the basic rights and freedoms of humans
View quoted post on X
1 replies 1 reposts 14 likes 1.4K views View on X
Laura Weidinger @weidingerlaura
Very proud of our 📣 new paper on Measuring Anthropomorphism in LLMs! This new multi-turn evaluation & large-scale human study led by @lujainmibrahim is a key step in better understanding what factors make people perceive LLMs as more "human-like". 🧵
Lujain Ibrahim Verified account @lujainmibrahim Feb 12, 2025
📣New preprint!📣We’ve long known humans tend to anthropomorphize computers. But with the rise of social AI applications, like AI companions, studying this is now more crucial than ever.

We introduce a new method for *empirically evaluating* anthropomorphic behaviors in LLMs🧵
View quoted post on X
1 replies 19 reposts 70 likes 7.1K views View on X
Laura Weidinger @weidingerlaura
Really enjoyed the discussion on using AI systems as content moderation tools - both the benefits and risks of what may (or will!) go wrong. Thanks for having me! @sfiscience Sign up to this series to attend future workshops.
Karine Mellata Verified account @karine_exe Nov 22, 2024
View quoted post on X