10% Time🔸

Well, Yoshua Bengio, Geoffrey Hinton and Stuart Russell actively support the bill. Hardly a outlandishly fringe AI doomsday cult I would say: computing.co.uk/news/4344580/b….

🚨With my PhD coming to an end, it's time for a life update! I've recently joined Qualcomm AI Research as a Senior Machine Learning Researcher. Looking forward to working with this great team!🧠✨

Using "speculative" as a pejorative is part of an anti-epistemic pattern that suppresses reasoning under uncertainty. If you disagree with someone's reasoning, just point out the flaw, or the premise you disagree with. If someone disparages an argument as "speculative", you

Oh man :( We tried really hard to neither over- nor underclaim the results in our communication, but, predictably, some people drastically overclaimed them, and then based on that, others concluded that there was nothing to be seen here (see examples in thread). So, let me try

Today, our Otto Barten⏸ and Tim Bakker🔸 are publishing a new oped in Dutch @Nrc Handelsblad, in which we argue that we should prepare for AGI. What are the risks that we see, and what do we propose to mitigate them? nrc.nl/nieuws/2025/02…

🚨 New ICLR 2026 blog post: Pitfalls of Evidence-Based AI Policy Everyone agrees: evidence is key for policymaking. But that doesn't mean we should postpone AI regulation. Instead of "Evidence-Based AI Policy," we need "Evidence-Seeking AI Policy." arxiv.org/abs/2502.09618…

If you pretend that xrisk from ASI misalignment is some novel, incredibly complex failure mode (instead of a simple extrapolation of established theories of incentive design), you blind people to the evidence for, and predictive power of, the theories that motivate the risk.

I recommend reading this paper ⬇️ It makes an evidence-based case that, without substantial effort to prevent it, AGIs trained like today’s top models could learn misaligned goals, hide them from their developers, and pursue them by seeking power.

Thread of photos from families in each quartile of income in the world: first photo is from the poorest 25%, last photo is richest 25%. Based on these photos, which income bracket are you in? First up: Toilets

Benchmarks saturate quickly, but don’t translate well to real-world impact. *Something* is going up very fast, but not clear what it means. Thus the wide range of expert opinion, from “superintelligence in a few years”, to “we’ve already hit a wall”. Our results shed some light:

Nick Nick, I respect you greatly but feel compelled to push back here as I did a year ago. Your scale and reasoning fundamentally misrepresents reality: When you evaluate "funding happy farm animals" at 10/10, you're ignoring crucial factors that make this position dubious. First,

I feel like a distressing amount of AI policy discourse stems from some people saying "without strong evidence for how risky powerful AI is, assume it's 100% safe and sit tight and assess", while I say "both are plausible, let's take actions that are reasonable in both worlds"

"How, exactly, could AI take over by 2027?" Introducing AI 2027: a deeply-researched scenario forecast I wrote alongside Scott Alexander, Eli Lifland, and Thomas Larsen

One sad fact about the current state of discourse is that whenever someone tries their hardest to do the impossible and predict the future, they're stigmatized by parts of the mainstream research community as doing "crackpottery". This causes reasonable people to not engage.

New essay exploring why experts so strongly disagree about existential risk from ASI, and why focusing on alignment as a primary goal may be a fundamental mistake

Come check out our TMLR-to-ICLR poster this afternoon "E-Valuating Classifier Two-Sample Tests". Time: 15-17:30 Where: Hall 3 + Hall 2B #437 Teodora Pandeva UvA AMLab #ICLR2025 #ML #Stats openreview.net/forum?id=dwFRo…

I'm increasingly concerned about the scenario of humans being gradually disempowered by AI, which could lead towards tyranny (if some small number of humans remain in charge) or even to humanity losing control of its future, all without a shot being fired. 1/2

.Ronny Chieng from The Daily Show talks about shrimp welfare. I unironically agree.