YouCongress Polls

Republicans keep trying to strip states of the ability to enact commonsense AI safeguards—at a time when there are no meaningful federal protections in place. President Trump's executive order is illegal coercion: it threatens states with costly laws... more Unverified source (2026)

The United States cannot remain competitive if developers, businesses, and users face fifty different legal regimes governing a general-purpose technology. [...] Congress should take this recommendation seriously and establish a light-touch national ... more Unverified source (2026)

The central risk is not that machines will outthink us, but that humans will abdicate responsibility. By framing AI as an autonomous decision-maker, especially in high-stakes domains, we risk normalizing the idea that accountability can be delegated ... more Unverified source (2026)

A "human in the loop" whose sole function is to approve a machine's actions is not a safeguard but a design failure. [...] In high-stakes environments, the illusion of human oversight is worse than no oversight at all. It creates confidence without c... more Unverified source (2026)

It is in the rational interest of various countries to make sure we end up with an international agreement. Unverified source (2026)

With nuclear weapons, [...] they can't [...] in secret, but with developing new types of AI, it's much easier [...] So it's not enough to have an agreement. Unverified source (2018)

That branch of AI is lethal. We've got to do something about that. Nobody should build it. And everyone needs to agree to that by treaty. Treaties are not perfect, but they are the best we have as humans. Unverified source (2026)

The idea of stopping or even substantially slowing the technology is fundamentally untenable. If one company does not build it, others will do so nearly as fast. [...] Even if all Western companies stopped their work on AI, authoritarian countries wo... more Unverified source (2026)

There are some digital products (say CSAM and AI letting terrorists make bioweapons) that I oppose regardless of whether they are open-source or not, but I'm overall supportive of open source, and you can easily verify that my MIT research group defa... more Unverified source (2026)

LLMs, such as GPT-4 and its successors, might provide dual-use information and thus remove some barriers encountered by historical biological weapons efforts. [...] BDTs may enable the creation of pandemic pathogens substantially worse than anything ... more Verified source (2023)

Since computational infrastructure is largely open-access, decentralized, and global, regulatory chokepoints are limited. Export controls may delay access to high-performance computing, but they are unlikely to prevent the use of open-source models f... more Verified source (2025)

For governments to allow private entities to essentially play Russian roulette with every human being on earth is, in my view, a total dereliction of duty. Unverified source (2026)

When 90% of the messaging is all around the end of the world and the pessimism...we're scaring people from making the investments in AI. It's not helpful. It's not helpful to people. It's not helpful to the industry. It's not helpful to society. It's... more Unverified source (2026)

I like the emphasis on transparent risk reporting and publicly verifiable safety roadmaps [but am] concerned that moving away from binary thresholds might enable a "frog-boiling" effect, where danger slowly ramps up without a single moment that sets ... more Unverified source (2026)

Those compliance costs are merely the beginning. The bill, if passed, would feed California regulators truckloads of company information that they will use to design a compliance industrial complex. Unverified source (2025)

Most risk management or governance looks at pre-deployment. But the majority of incidents we have to worry about cannot be reliably predicted before deployment. [...] We need standardised definitions of incidents and accidents. Unverified source (2026)

If you have to report to 50 different state regulators at 50 different times with 50 different definitions, it's extremely onerous. And it's going to slow down innovation, and it's going to hinder our progress in the AI race. Unverified source (2025)

I'm now very confident that it is possible to build AI systems that don't have hidden goals, hidden agendas. [...] A Scientist AI would be trained to give truthful answers based on transparent, probabilistic reasoning. Unverified source (2026)

[...] principles requiring that AI be explicable are misguided. We should be deciding which decisions require explanations. Automation is still an option; Unverified source (2019)

It is somewhat awkward to say this as the CEO of an AI company, but I think the next tier of risk is actually AI companies themselves. [...] AI companies control massive data centers, train the most advanced models, and possess unmatched expertise in... more Unverified source (2026)

Policy and safeguards must evolve just as swiftly [...] requiring safety assurances for powerful open releases before the window for meaningful intervention closes. Unverified source

I do not believe this is the best approach to protecting the public from real threats posed by the technology. Unverified source (2024)

Open source models are technically exempt [...] based on the assumption that the system is sufficiently well documented to convey its potential risks. Unverified source (2024)

We commend the BIS rule’s inclusion of a provision which requires the reporting [...] computing power greater than 10^{26} integer or floating-point operations. Unverified source (2024)

According to the proposed rule, a covered US person will be required to provide quarterly reports to BIS if the US person “engages in, or plans, within six months, to engage in ‘applicable activities.’” “Applicable activities” include: “Conducting a... more Unverified source (2024)

Once AI reaches human-level capabilities, we will be able to use it to improve itself in a self-referential way. I personally believe that if we can reach AGI, we will reach ASI shortly, maybe a few years after that. Unverified source

Until some company or scientist says ‘Here’s the proof! We can definitely have a safety mechanism that can scale to any level of intelligence,’ I don’t think we should be developing those general superintelligences. We can get most of the benefits w... more Unverified source (2024)

My own opinion is that of course this is possible in principle. I would never have started working on Artificial Intelligence if I did not believe that. [...] Even if it is possible I personally think we are far, far further away from understanding ... more Unverified source

We will refer to such a lab as the Global Public Benefit AI Lab (or "Lab"). The Lab will be an open, public-private, democratically-governed consortium, aimed at achieving and sustaining a solid global leadership or co-leadership in human-controllab... more Unverified source

To mitigate the “race to God-like AI,” Ian Hogarth—chair of the U.K. AI Safety Institute—proposed an “Island model,” in which a joint international lab performs research on superintelligence in a highly secure facility. An essential part of this pro... more Unverified source (2024)

So the the prototypical situation I'm imagining here is you know there's a kind of one ai project, which is you know somewhat ahead of the others, and maybe it it goes through intelligence explosion, whereas which by which I mean kind of AI can autom... more Unverified source

The thought of open source AGI being released before we have worked out how to regulate these very powerful AI systems is really very scary, Unverified source (2024)

These studies, on their own, are insufficient evidence to demonstrate increased marginal societal risk from open foundation models. Unverified source (2024)

You can see exactly what is being done, because all the code is published so it can be publicly audited. Unverified source (2025)