Coalition for a Baruch Plan for AI (@BaruchPlanForAI) header
Coalition for a Baruch Plan for AI (@BaruchPlanForAI) profile picture

A coalition of NGOs to foster a uniquely timely, bold and effective treaty-making process to build the global federal organization that we need to master AI.

Geneva, Switzerland
Joined September 2024
Like it or not, given the shocking acceleration of AI risk, the future of humanity rests on Trump and Xi co-leading an extremely bold global AI treaty that will reliably prevent Superintelligence, reduce concentration of power and wealth and realize AI's astounding potential: grok.com/imagine/post/9d51e2…
🤖 Made with AI
1
1
90
🇨🇳 Yesterday at Davos, China's Vice-Premier He Lifeng called for global AI governance cooperation. Is this genuine—or just strategic positioning? We analyzed 2 years of evidence. Our take: The evidence suggests China would genuinely engage in a comprehensive AI treaty. The chokepoint isn't Beijing's willingness. It's whether Trump can be persuaded to co-lead. 🔗 cbpai.org/blog-1/is-chinas-c… #AIgovernance #baruchplanforai #Superintelligence #ASI
2
87
Apparently, the quoted post by Prof Bengio has been deleted (here is a snapshot of it). We asked him why. We'll post below any reply we receive from him. Our guess is that, likely, he needs more time to review our just published 90-page Case for a Coalition for a Baruch Plan for AI before endorsing us via social media, which is a very big deal in this day and age (Bengio is new to X, as we are) - especially considering how geopolitically disruptive, outside the Overton Window, and opposed to *current* US/UK AI intl’ policy, our initiative is. But the reasons may go much deeper, having primarily to do with the extremely critical role that influential AI scientists find themselves in, as AI becomes the main instrument in an accelerating arms and dominance race among superpowers. Consider that Bengio leads several major and praiseworthy AI initiatives of the UK gov, so he has to be careful endorsing initiatives that directly counter the policy of his employer. In fact, these are largely the same reasons why at least the 7 leading AI experts or AI labs policy chiefs who supported in the past the idea of a Baruch Plan for AI (link above in my post), have not yet come out strongly and decisively about it, or joined or supported our open coalition publicly. These same dynamics were at play in 1945-6 for leading atomic scientists, as we argue in pages 28-29 of our Case for a Coalition for a Baruch Plan for AI paper, (pages 28-29): cbpai.org/case-for-a-coaliti… "Yet many of the most authoritative AI scientists - such as Yoshua Bengio, Paul Christiano, and Ian Hogarth - have assumed crucial formal roles in the work of key AI states. As it happened in 1945-46 for many state-employed top nuclear scientists, they are in a very delicate position that could make it harder for them to too openly embrace strong global governance of AI while the states are mostly set in an arms race and alliance-based mostly unilaterally lead approach to international governance of AI. Others who remained on the outside such as Neil Bohr and Albert Einstein were able more freely to advocate for a stronger global governance of AI and for comment-specific plans. Still, Oppenheimer, while on the inside (and doubted to be a communist) became decisive in writing the Acheson-Lilienthal Report that convinced President Truman to ask Bernard Baruch to turn that into a treaty proposal." This is not to mention how it is an historical fact that influential scientists have been and will be bought with money and prestige, pressured or threatened (or attempted to do so) when immense geopolitical interests are at play. Or the tight and largely arbitrary power of a few state officials (opposed to intl governance) on which classified information is shared with whom, at what clearance level or publicly released, and when it is (as Oppenheimer learned, through the activities of General Groves, teaching many others). Still, the courage of atomic scientists and the fast raising evidence of the impossibility of a nation alone to prevent catastrophic nuclear weapons proliferation, made independent and dependent top scientists lead Truman and Stalin to formally propose the Baruch and Gromyko Plans. Yet, the work and courage by such nuclear scientists was not sufficient to make so that an agreement on a middle ground between a Baruch and Gromyko plan could be achieved. A failed approach at treaty-making by Truman and Stalin made so that those top state-employed nuclear scientists (who had classified information clearance) were largely prevented to have substantive dialogue with their counterparts in the months before the presentation of the Baruch Plan and Gromyko Plans, which could have created an agreeable common ground. Hopefully, we’ll do better with AI, as we must. Last week, we were asked by the prestigious Bulletin of Atomic Scientists for a detailed 3-4000 words Op-ed on these issues. We’ll post in this comment thread such an op-ed as published.
3
1
125
Existential risks loom as AI advances rapidly. The Coalition for a Baruch Plan for AI proposes bold global governance to ensure safety & benefits for all. Learn more in our 8-minute introductory video: youtu.be/FqyqFlpv1A8
1
1
7
626
We published today the Case for a Coalition for a Baruch Plan for AI (v.1). 22 global experts contributed to this exploration of the urgent need for a democratic global governance to AI risks, inspired by the 1946 Baruch Plan for nuclear tech: cbpai.org/publications
1
2
4
150
13/ Join us in humanity's greatest challenge. Sign our Open Call, become a partner, or support our mission: cbpai.org/open-call-v1 Together, we can ensure AI becomes humanity's greatest invention rather than its worst and last.
1
4
73
As we end 2024, humanity faces its greatest challenge yet: ensuring AI remains safe and beneficial for all. The Coalition for a Baruch Plan for AI launches an urgent call to states to create a bold global governance of AI through unprecedented cooperation. cbpai.org/open-call-v1
3
1
8
1,432
Today, we are pleased to announce the Coalition for a Baruch Plan for AI to unite the world to ensure a safe and fair governance of AI, inspired by bold initiatives from our past. cbpai.org @BaruchPlanForAI #AIgovernance #AIsafety
1
1
4
415