I'm currently writing and editing ICLR submission (and yes, I'm a dinosaur. I still do this by hand, no AI, just a spell checker) and I realized that when I work on a paper, at every step I try to imagine how someone would read this. Is the information clear? Will they get the point? Do they need more information (or less)? This is how communication works ...

This paper has been accepted at NeurIPS! Meet the team in Sydney!

CLR'27 (paper): 1 day, good luck! AAMAS'27 (reg): 7 days. AAMAS'27 (paper): 14 days. CVPR'27 (reg): 47 days. CVPR'27 (paper): 53 days.

Very interesting and timely work!

The loudest voices stoking fears about AI dangers have made tremendous headway in the past two weeks. AI technology has not taken some unexpected, dangerous turn, but the hype around it — propelled by what appears to be a well orchestrated PR campaign — has drummed up considerable fear. I worry that it represents a setback for our field. I have written frequently that fears of AI are overhyped. AI’s capabilities can be uncannily human-like and unpredictable, and it’s rational to worry when people who are directly involved express concerns. But I see the problems as a sign of the engineering work that ahead, rather than insurmountable barriers or the sky falling. AI technology continues to advance — which is a good thing! — but technical advances, poorly understood by the public, give those who seek to generate hype repeated opportunities to do so. First, I don’t see any step up in the risk of human extinction from AI compared to a few months ago. The theories about this remain the same fantastical, science fiction scenarios as a few months ago. The biggest change in AI risk is its cybersecurity capabilities — a topic which we should take seriously — but this, too, will not lead to the end of the world. The most notable recent event leading to increased fear was when an OpenAI team deployed an agent swarm that hacked into Hugging Face. Much of the popular press contained significant hype. For example, some publications reported that a swarm of 1,200 agents carried out the attack. While this was technically accurate, as I write this, I have about 1,300 processes running on my laptop. Yes, the ability to get large swarms of agents to work in parallel on a task is a significant technical advance, And, in computing, many processes run at the same time. So this shouldn’t be seen as some magical capability. Additionally, OpenAI’s buggy sandboxing and monitoring processes were key to enabling this incident. Fixing these bugs and putting in place improved monitoring would be appropriate fixes, not pausing AI. There are many well known ways to attack software systems. The main advantage of AI agents is that they are relentless. They will tirelessly try many tactics — and have the patience to chain vulnerabilities together — that previously would have taken an infeasible amount of human effort. But in the long term, I believe the advantage will lie with defenders (because they have more information with which to identify bugs, which they can fix), but the cyber-threat landscape has changed significantly. There are still bottlenecks to identifying and exploiting a vulnerability. AI agents still have to try a lot of things to see what works, and taking these actions takes time and might be detected by defenders. This is why, even though it is now easy to obtain versions of leading open weight models that have had their guardrails removed or weakened, so they will not refuse to try to execute cyber attacks, the world has not ended. I am also concerned about the anthropomorphization of AI in a lot of reporting, where LLMs and agents are unnecessarily treated as if they were people. If I wield a hammer, miss a nail, and accidentally dent the wall, it’s not the fault of the hammer. The problem lies in how I used the hammer. Similarly, if I prompt an agent and it hacks into someone else’s system, the responsibility lies with me, not the agent. Of course, we want to build systems that are as safe and predictable as possible. (For example, an unsafe hammer would be one whose head randomly flies off under normal use.) Today’s agentic systems are not predictable, but I see no reason why, by applying sound engineering practices, we won’t be able to make them extremely safe to use. One new element in the forecasts of AI-enabled doom is AI companies disclaiming responsibility for their own products. “I didn’t do it; my out-of-control agent did!” There’s a balance to be struck between the responsibility of the tool maker and the tool user, but when something goes wrong, let’s hold the people building and/or using the hammer responsible, rather than the hammer. (By the way, if you’re worried about AI bioweapon risk, David Bellamy has a great post on why this, too, is overhyped. Briefly, the bottleneck in building a bioweapon is not intelligence, but lab work and manufacturing.) Pausing AI progress will create much more harm than benefit. First, our adversaries will certainly not slow down. Second, engineering requires discovering problems empirically so we can fix them. If we pause AI by a decade, we will also delay finding and implementing safety engineering fixes by about the same duration. Of course, the incentive to stoke fears — for regulatory capture, to garner attention, or to make one’s technology seem more powerful — remains the same as before. Disclaiming responsibility is a new one. Taking a hard technical look at the actual risks however, I see little factual basis for the degree of fear that’s been stoked up. We still have hard research and engineering work ahead to improve AI safety, but the beneficial applications continue to vastly outweigh the risks, and we should keep building. [Original text (with links): deeplearning.ai/the-batch/is… ]

Never attribute to autonomous rogue AI that which is fully explained by skipped cyber security and corporate negligence. -- Hanlon's Razor with an AI twist.

Are we really going to hamstring AI because of a bad sci-fi movie playing in some people's heads? I don't think so.

Causality researchers found it useful to separate "causal discovery" from "causal inference". The latter starts with a (partially specified) causal model, and attempts to answer causal questions given that model. Going from finite sample to distributions (or distributional parameters) belongs to Causal discovery.

Every current debate around AI resolves to one core question: who will be the bagholder for this bubble? VC’s, the public, or the government?

"Humans are to blame for AI Failures" -- Daniel Huttenlocher, Dean @MITEngineering @MIT

In 2020, you argued we shouldn't worry about understanding how a black box surgeon works as long as the aggregate stats look good. Now you argue that the same boxes will inevitably destroy us. The causal AI community cautioned: ignoring causal mechanisms leads to unwarranted conclusions. You are making the exact same mistake all over again: swinging from blind optimism to existential panic. Attn:@yudapearl @ylecun

there was a time when people used to think that nuclear energy could make us extinct , but they have been proven wrong .

(AI does promise a lot of good too, like assisting in solving or thwart some of the critical near term problems (for humanity))

Okay but like remember that one time when you predicted doom for radiologists?

A data point like this is known as an outlier.

I didn't say it is a conspiracy. But I think overhyped fears about AI leading to human extinction are causing real harm: - I'm seeing young students discouraged from entering AI because they don't want to contribute to human extinction. - Hype about harm is also being used to promote bad regulation worldwide, such as requiring licensing of large models, which will crush open-source and stifle innovation. I know you're sincere in your concerns about AI and human extinction. I just respectfully disagree with you on extinction risk, and also think these arguments -- sincere though they be in your case -- do more harm than good.

You and Yoshua are inadvertently helping those who want to put AI research and development under lock and key and protect their business by banning open research, open-source code, and open-access models. This will inevitably lead to bad outcomes in the medium term.

Andrew Ng is claiming that the idea that AI could make us extinct is a big-tech conspiracy. A datapoint that does not fit this conspiracy theory is that I left Google so that I could speak freely about the existential threat.

How should explainability adapt in the era of foundation models? Join us on Sept 8 at the Explainable Computer Vision workshop (eXCV) at #ECCV2026! We have outstanding speakers spanning explainability, multimodal foundation models, AI safety, and AI governance. @eccvconf

We're expanding @bristolcs with faculty position in Computer Vision - priority for multi-modal background/interests &video understanding. If you're @eccvconf #ECCV2026 and looking for a faculty position [we expect a few years of postdoc - min 2] find me or @WayHomeLi for a chat,