New working paper on instrumental convergence at the GPI working paper series globalprioritiesinstitute.org/what-power-see…

Happy to share a CFP for the upcoming "Conscious AI? Functionalism and its Alternatives" workshop, with Anil Seth (Anil Seth) and Kathinka Evers as keynote speakers. The workshop is scheduled for June 2–3, 2025 in Prague. More info here: emw.flu.cas.cz

I find it so frustrating to read Vox articles like this: "Optimizing for beneficent impact doesn't maximize my own happy feelings, so it must be bad. Indeed, it must be motivated by a pathological desire for control - what else could explain why people want fewer kids to die?

nytimes.com/2025/04/03/tec…

Do you have a research interest in how AI can harm or benefit non-human animals? I'm recruiting a Research Officer to drive forward the "Animals and AI" priority area of The Jeremy Coller Centre for Animal Sentience at the LSE. Please apply by 6 May! 🐖🐔🐟jobs.lse.ac.uk/Vacancies/W/67…

In the long-run, this is far more important news than the stock bounce today. Few policy questions are as clear cut. If this continues, the US will essentially be forfeiting its AI leadership in order for NVIDIA to make slightly more profit this year. x.com/hamandcheese/s…

Call for Papers Bi-annual Conference on the Philosophy of Artificial Intelligence (PhAI 2025) Vrije Universiteit Amsterdam | 23 + 24 October 2025 (deadline 30.06.25) philevents.org/event/show/134… (also pt-ai.org) please RT

x.com/i/article/1910…

Very happy to announce that our paper “Sensory Horizons and the Functions of Conscious Vision” is now out as a target article in BBS! Steve Fleming and I present a new theory of the evolution and functions of visual consciousness. Article here: doi.org/10.1017/S01405…. A thread🧵

Matthew Barnett Suppose we automate all human knowledge workers completely. The only economic power a person has is derived from financial and social capital. What happens to those humans who lack either? Are they serfs forever? UBI is a bad answer because it can be taken away as easily as it

This is such zinger. Career advice from Rutger Bregman 👏👏👏

"we do not wish to advance the rate of AI capabilities progress" 😇 - Anthropic like 2 years ago "We want Claude n to build Claude n+1" - Anthropic today

There is zero evidence that China is seriously pursuing superintelligence, or is pursuing the global dominance described here. Meta is paying these organisations to scare American policymakers into opposing regulation of AI. This is commercial propaganda. papers.ssrn.com/sol3/papers.cf…

In our new paper (open access in Phil. Studies), Albert Newen and me investigate the potential and limitations of multidimensional models of animal (and potentially AI) minds, many of which have been recently developed: link.springer.com/article/10.100…

🚨🚨 Excited to share a new paper led by Haiwen Li with the Community Notes team! LLMs will reshape the information ecosystem. Community Notes offers a promising model for keeping human judgment central but it's an open question how to best integrate LLMs. Thread👇

The “Manhattan Project” framing of AI alignment--as a binary, technical challenge that can be solved such that AI takeover is averted--is misleading. It's neither clear-cut nor fully operationalizable. New paper with Leonard Dung in Mind and Language: onlinelibrary.wiley.com/doi/10.1111/mi…

Great paper by Simon Friederich and Leonard Dung! We agree that "solving the alignment problem" is not obviously the solution to reduce existential risk. We also agree that we should not frame xrisk reduction in terms of a Manhattan Project. "One possible outcome is to conclude

Great discussion in Matthew Yglesias 's mailbag today about loss of control risk.

The Global Priorities Institute has now shut its doors. For an overview of what it achieved, see: globalprioritiesinstitute.org

Can benefits for species with small welfare ranges outweigh significant human goods? My new paper explores this issue and argues that the answer is plausibly ‘yes’. Now available online first: journals.publishing.umich.edu/jpe/news/207/ For the preprint-pdf see philpapers.org/rec/LOHTMI-2