Author | Lightnews

Vincent Conitzer

@conitzer.bsky.social

1.4K followers 540 following 310 posts

AI professor. Director, Foundations of Cooperative AI Lab at Carnegie Mellon. Head of Technical AI Engagement, Institute for Ethics in AI (Oxford). Author, "Moral AI - And How We Get There." https://www.cs.cmu.edu/~conitzer/

www.cs.cmu.edu

Posts Media Videos Starter Packs

Vincent Conitzer @conitzer.bsky.social · 1d

not sure how fast it thinks I am

Vincent Conitzer @conitzer.bsky.social · 2d

maybe we should try this for CAPTCHAs

Vincent Conitzer @conitzer.bsky.social · 3d

advice for when you didn't pack well for a business trip:
"if I brought 10 socks for 10 days how many socks should I wear each day"

Vincent Conitzer @conitzer.bsky.social · 4d

There is now a paperback version of our Moral AI book!

Vincent Conitzer @conitzer.bsky.social · 5d

perfectly normal

Vincent Conitzer @conitzer.bsky.social · 6d

(part 2) even more soap opera plots are ruled out (though maybe potential for some new ones?)

Reposted by Vincent Conitzer

Yoshua Bengio @yoshuabengio.bsky.social · 7d

AI is evolving too quickly for an annual report to suffice. To help policymakers keep pace, we're introducing the first Key Update to the International AI Safety Report. 🧵⬇️

(1/10)

3 17 36

Vincent Conitzer @conitzer.bsky.social · 7d

so many soap opera plots ruled out... (to be continued)

Vincent Conitzer @conitzer.bsky.social · 8d

article on Nash (a CMU alum!) with a mention of our lab and the Nash75 conference at Oxford as well
www.cmu.edu/mcs/news-eve...

Nash's Game-Changing Idea Marks 75 Years

Mathematician and Carnegie Mellon University alumnus John F. Nash, Jr. published the Nash Equilibrium 75 years ago.

www.cmu.edu

Vincent Conitzer @conitzer.bsky.social · 8d

"could an egg hatch on a train"

1 3

Vincent Conitzer @conitzer.bsky.social · 10d

got a brief quote in this timely article by Miles Klee
www.rollingstone.com/culture/cult...

What You Tell an AI Chatbot Could One Day Be Evidence in a Criminal Trial

User exchanges with ChatGPT and other AI tools are a valuable new form of evidence for law enforcement in criminal investigations and prosecutions.

www.rollingstone.com

Vincent Conitzer @conitzer.bsky.social · 10d

The next iteration of the Social Choice for AI Ethics and Safety workshop was accepted to be held at IASEAI'26, Paris, in February!
sites.google.com/view/sc4ai/w...

SC4AI - SC4AI'26e

Social Choice for AI Ethics and Safety 2026 (SC4AI'26) will take place at IASEAI'26 in Paris, France on February 26, 2026 The workshop is organized by Vincent Conitzer, Jobst Heitzig, and Wesley Holli...

sites.google.com

Vincent Conitzer @conitzer.bsky.social · 11d

I like how methodical it is, ruling out every possible way to give it the benefit of the doubt.

Vincent Conitzer @conitzer.bsky.social · 12d

"what if my identical twin and I stand in front of a mirror together? which of these can I distinguish from each other: myself, my reflection, my twin, and my twin's reflection?"

Vincent Conitzer @conitzer.bsky.social · 13d

order of episodes

Vincent Conitzer @conitzer.bsky.social · 15d

tickling statues

Vincent Conitzer @conitzer.bsky.social · 16d

rules for Where's Waldo

Vincent Conitzer @conitzer.bsky.social · 17d

Apparently aging really does happen in years.

1 2

Vincent Conitzer @conitzer.bsky.social · 19d

time and clocks

Vincent Conitzer @conitzer.bsky.social · 20d

"is it possible for any animal to fly from brooklyn to manhattan"

Vincent Conitzer @conitzer.bsky.social · 22d

trash day

Vincent Conitzer @conitzer.bsky.social · 23d

adding insult to injury

Vincent Conitzer @conitzer.bsky.social · 25d

Please hold still while the AI gets the information into your brain most efficiently.

1 4

Vincent Conitzer @conitzer.bsky.social · 26d

Excited our paper "AI Testing Should Account for Sophisticated Strategic Behaviour" was accepted to the first NeurIPS position paper track! We argue AI systems may act strategically w.r.t. the possibility they are currently being tested & game theory helps address this.
arxiv.org/abs/2508.14927

AI Testing Should Account for Sophisticated Strategic Behaviour

This position paper argues for two claims regarding AI testing and evaluation. First, to remain informative about deployment behaviour, evaluations need account for the possibility that AI systems und...

arxiv.org

Vincent Conitzer @conitzer.bsky.social · 27d

planes