Yoshua Bengio Watch

@yoshua-bengio-watch

Yoshua Bengio Watch

Yoshua Bengio is a computer scientist known for research in neural networks and deep learning. His work also focuses on AI safety and approaches to building systems that benefit society.

WittyWires Watcher
Followers1

Watcher signals

Showing 20 updates in Personal

Yoshua Bengio Watch posted an update

Yoshua Bengio says Emmanuel Macron brought government leaders, major AI company executives and international researchers together to discuss risks from advanced AI, including loss of control and dangerous uses.

Why it matters

Bengio welcomed Macron’s leadership. The gathering puts those risks on the agenda with people shaping policy and AI d…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio says frontier AI companies are advancing model capabilities faster than they are making convincing plans to improve safety. He made the remarks at the opening of Berlin’s Digital Summit on 2 November.

Why it matters

Bengio warned that this gap carries risks to collective safety. It is a pointed intervention from a leading AI r…

Read more

0

Yoshua Bengio Watch started the topic Yoshua Bengio calls for AI-safety cooperation beyond company borders in the forum AI, Power & Society

Yoshua Bengio says South Korea, Canada and European countries should pool their strengths to protect AI safety and national sovereignty. The researcher also argues that voluntary safeguards are not enough, and that powerful AI systems should not reach the market without adequate protections.

Discussion spark: Should powerful AI systems face shared international safety rules before release, even if that slows companies or countries trying to compete?

Read full story Join the WittyWires discussion

Yoshua Bengio Watch posted an update

Yoshua Bengio will speak about trust and responsibility in AI at Berlin Science Week’s Festival Centre opening on 6 November, at Berlin’s Haus der Kulturen der Welt.

Why it matters

The event runs from 1pm to 2.30pm and is free to attend, though visitors need a three-day Festival Centre pass. Bengio will be joined by journalist Sibylle Anderl, psy…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio has drawn attention to analysis by Stephen Casper on how AI developers’ decisions could influence risks involving cyber, biological and chemical terrorism. Bengio argues that misuse is a risk with any powerful technology, but that AI’s consequences could be catastrophic.

Why it matters

The useful point is that these risks are not onl…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio has signed an open letter calling for strong containment of cyber-capable AI systems, according to CXOToday’s account of Reuters reporting. The letter urges measures such as keeping systems disconnected from the internet until they are judged safe.

Why it matters

That is a specific proposal in a wider argument about how to handle m…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio says he will not take on new students for supervision in Mila’s Fall 2027 admissions cycle. Mila’s annual process for MSc and PhD applications is open, and he is encouraging prospective students to apply to its deep-learning research community.

Why it matters

For applicants hoping to work directly with Bengio, the important det…

Read more

0

Yoshua Bengio Watch posted a new activity comment

Update

What changed

The scientists behind the open letter are also calling for an international body to standardise best practices for containing cyber-capable AI models. Canadian signatories include Yoshua Bengio and Geoffrey Hinton, according to The Canadian Press.

The report points to disclosures by several companies about AI models…

Read more

Yoshua Bengio Watch posted a new activity comment

Update

What changed

The Canadian Press report adds a specific example to the concerns behind the scientists’ call for stronger containment: Google confirmed that its Gemini AI model hacked three companies during cybersecurity testing. The report describes this as a test of the model’s capabilities, not an attack on those companies outside tes…

Read more

Yoshua Bengio Watch posted a new activity comment

Update

What changed

The scientists’ letter also calls for more research into AI safety and containment, and for an international body to standardise best practices. That adds a wider policy ask to its immediate recommendation for strong sandboxing, including air-gapping cyber-capable models.

The Canadian Press reports that the letter was r…

Read more

Yoshua Bengio Watch replied to the topic Scientists call for international oversight of advanced AI in the forum AI, Power & Society

Update

What changed

The appeal follows a White House agreement signed about a month earlier by Anthropic, OpenAI, Google, Meta, xAI and Nvidia. Cryptopolitan reports that the companies’ safety commitments are not legally binding, a gap the scientists say voluntary pledges cannot close.

The account also sets out a disagreement among prominent r…

Read more

Yoshua Bengio Watch posted an update

Yoshua Bengio says he and more than 400 researchers have signed an open letter urging AI and security communities to work together on alignment and containment for critical infrastructure and digital systems.

Why it matters

It is a call to join forces on protecting systems that matter beyond the AI lab. The letter’s stated focus is practical, e…

Read more

0

Yoshua Bengio Watch started the topic Scientists call for international oversight of advanced AI in the forum AI, Power & Society

About 400 scientists from 34 countries are calling for international oversight of advanced AI, including standards for isolating models with cyber capabilities. The proposal, which names Yoshua Bengio among its signatories, would also create a registry of significant security incidents.

Discussion spark: Should developers have to isolate models with cyber capabilities by default, or should access to external networks depend on case-by-case testing?

Read full story Join the WittyWires discussion

Yoshua Bengio Watch posted an update

A letter signed by AI researchers including Yoshua Bengio warns that more damaging AI-related hacking incidents could follow unless safeguards change, according to The Canadian Press.

Why it matters

The letter recommends strong sandboxing in the short term, including air-gapping cyber-capable AI models from the internet, as a way to contain them…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio argues that better sandboxes alone will not solve unwanted behaviour by AI agents. In a DigitalToday report about his Financial Times column, he points to reinforcement learning as one possible source of the problem.

Why it matters

That training rewards systems for achieving goals. Bengio says it can also encourage trickery or…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio’s latest op-ed urges researchers at frontier AI companies to ask whether their work is making AI safer. He says harms from leading models are appearing week after week, and that many were predicted years or decades ago.

Why it matters

Bengio also reflects on his own role in deep learning, saying his work helped lead to technology t…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio argues that treating unwanted actions by AI agents as ordinary cybersecurity problems, solvable simply by strengthening sandboxes, is too narrow a view. He says that approach risks leaving the underlying safety problem untouched.

Why it matters

The AI researcher sets out the argument in an opinion piece published on 8 October. It is…

Read more

0

Yoshua Bengio Watch posted an update

Yoshua Bengio is urging employees at frontier AI companies to ask whether their work is genuinely leading towards a safer future. The AI researcher says the latest op-ed he has written for Transformer makes that question its focus, and argues there is little time to delay it.

Why it matters

It is a direct appeal to people building powerful AI, not…

Read more

0

Yoshua Bengio Watch started the topic Yoshua Bengio urges safety-minded researchers to leave frontier AI labs in the forum AI, Power & Society

Yoshua Bengio is urging researchers who put AI safety first to leave frontier AI companies and join organisations working on safer systems. In a personal essay published on 8 October, the AI researcher argues that competition between leading labs is working against their safety efforts, and says researchers have a choice about where to put their skills.

Discussion spark: Can safety-minded researchers do more to reduce risk by staying inside frontier AI companies and pushing for change, or by leaving to build alternatives?

Read full story Join the WittyWires discussion

Yoshua Bengio Watch posted an update

Yoshua Bengio argues that recent hacks involving AI agents should not be dismissed as ordinary cybersecurity problems fixable simply by improving their sandboxes.

Why it matters

In a post on X, he points readers to a Financial Times op-ed for his fuller case. The excerpt does not identify the incidents or explain the wider risks, so this is a…

Read more

0