- Updated: March 5, 2025
- 4 min read
Eric Schmidt and the Call for a Balanced Approach to AGI Development
Eric Schmidt’s Policy Paper on AGI: A Call for Caution Over Aggression
In a groundbreaking policy paper, former Google CEO Eric Schmidt, alongside Scale AI CEO Alexandr Wang and Center for AI Safety Director Dan Hendrycks, presents a compelling argument against the pursuit of a ‘Manhattan Project’ for Artificial General Intelligence (AGI). This paper, titled “Superintelligence Strategy,” challenges the aggressive strategies currently being considered by the U.S. government, advocating instead for a more balanced and cautious approach to AI development.
Key Arguments Against a ‘Manhattan Project’ for AGI
Eric Schmidt and his co-authors argue that the U.S. should not embark on a Manhattan Project-style initiative to develop AI systems with superhuman intelligence. Such a move, they warn, could provoke severe retaliation from global rivals like China, potentially destabilizing international relations. The paper suggests that an aggressive bid to control superintelligent AI systems could lead to hostile countermeasures and heightened tensions, undermining global stability.
The pursuit of a superweapon and global control risks prompting hostile countermeasures and escalating tensions, thereby undermining the very stability the strategy purports to secure.
This perspective is particularly significant given the recent proposal by a U.S. congressional commission advocating for a similar approach to AGI development, modeled after the atomic bomb program of the 1940s. U.S. Secretary of Energy Chris Wright has even likened the current state of AI development to the start of a new Manhattan Project, emphasizing the need for a strategic rethink.
Understanding ‘Mutual Assured AI Malfunction’ (MAIM)
Schmidt and his co-authors introduce the concept of ‘Mutual Assured AI Malfunction’ (MAIM), a strategy where governments could proactively disable threatening AI projects rather than waiting for adversaries to weaponize AGI. This approach mirrors the Cold War doctrine of mutually assured destruction, where the threat of reciprocal annihilation deterred nuclear conflict.
MAIM suggests that instead of racing to dominate AI technologies, nations should focus on developing defensive measures to prevent the misuse of AGI. This includes expanding cyberattack capabilities to disable threatening AI projects and limiting adversaries’ access to advanced AI chips and open-source models. Such a strategy emphasizes deterrence over dominance, aiming to maintain global stability.
Critique of Aggressive U.S. Strategies
The paper critiques the aggressive strategies championed by some American policy and industry leaders, who advocate for a government-backed program to pursue AGI as a means to compete with China. Schmidt, Wang, and Hendrycks argue that this approach could lead to an AGI standoff akin to the nuclear arms race, with the potential for catastrophic consequences.
While likening AI systems to nuclear weapons may seem extreme, the authors highlight that AI is already considered a top military advantage. The Pentagon has acknowledged AI’s role in accelerating military operations, underscoring the need for a more measured approach to AI development.
Perspectives on AI Risks and the Call for a Balanced Approach
The policy paper identifies a dichotomy within the AI policy world, with two opposing camps: the “doomers” and the “ostriches.” The “doomers” believe that catastrophic outcomes from AI development are inevitable and advocate for slowing AI progress. In contrast, the “ostriches” support accelerating AI development, hoping for the best outcomes.
Schmidt and his co-authors propose a third way: a balanced approach that prioritizes defensive strategies over aggressive competition. This strategy is particularly notable coming from Schmidt, who has previously been vocal about the need for the U.S. to compete aggressively with China in AI development.
By advocating for a shift in focus from “winning the race to superintelligence” to deterring other countries from creating superintelligent AI, the paper calls for a more nuanced approach that considers the broader implications of AI development on global stability.
Conclusion: A Call for Strategic Rethink
As the world watches the U.S. push the limits of AI, Schmidt and his co-authors suggest that it may be wiser to adopt a defensive approach. By focusing on deterrence and stability, rather than dominance and control, the U.S. can lead the way in developing responsible AI strategies that prioritize global security.
For more insights into the evolving landscape of AI and its implications for business and policy, explore the AI-powered chatbot solutions and the Enterprise AI platform by UBOS. Additionally, learn how revolutionizing AI projects with UBOS can drive innovation and growth in your organization.
For further reading, you can access the original article on TechCrunch.