https://www.nationalsecurity.ai/ [679d4ad9c7] Download PDF Standard Version 10 pages Expert Version 32 pages Superintelligence Strategy Dan Hendrycks , Eric Schmidt , Alexandr Wang Abstract Rapid advances in AI are beginning to reshape national security. Destabilizing AI developments could rupture the balance of power and raise the odds of great-power conflict, while widespread proliferation of capable AI hackers and virologists would lower barriers for rogue actors to cause catastrophe. Superintelligence--AI vastly better than humans at nearly all cognitive tasks--is now anticipated by AI researchers. Just as nations once developed nuclear strategies to secure their survival, we now need a coherent superintelligence strategy to navigate a new period of transformative change. We introduce the concept of Mutual Assured AI Malfunction (MAIM): a deterrence regime resembling nuclear mutual assured destruction (MAD) where any state's aggressive bid for unilateral AI dominance is met with preventive sabotage by rivals. Given the relative ease of sabotaging a destabilizing AI project--through interventions ranging from covert cyberattacks to potential kinetic strikes on datacenters--MAIM already describes the strategic picture AI superpowers find themselves in. Alongside this, states can engage in nonproliferation to rogue actors to keep weaponizable AI capabilities out of their hands, and they can increase their competitiveness by bolstering their economies and militaries through AI. Taken together, the three-part framework of deterrence, nonproliferation, and competitiveness outlines a robust strategy to superintelligence in the years ahead. [67c872fb73] Start reading Authors [67a057fcb988c] Dan Hendrycks Director of the Center for AI Safety [67a058a1e1ba1] Eric Schmidt Former CEO and Chairman of Google, KBE [67ad815adaa7a] Alexandr Wang Founder and CEO of Scale AI Audio Releases Standard Version: Expert Version: Media & Press * No Priors Podcast National Security Strategy and AI Evals on the Eve of Superintelligence with Dan Hendrycks * Daily Caller Defense, Tech Leaders Warn AI Could Unleash A New Cold War * Nonzero Mutual Assured AI Malfunction | Robert Wright & Dan Hendrycks [679d4ad9c7] Table of contents Download Standard Version (10 page PDF) Download Expert Version (32 page PDF) 1 Executive Summary 2 Introduction 3 AI Is Pivotal for National Security 3.1 Strategic Competition 3.1.1 Strategic Competition Shifting Basis of Economic Power 3.1.2 Strategic Competition Destabilization Through Superweapons 3.2 Terrorism Terrorism 3.2.1 Terrorism Bioterrorism 3.2.2 Terrorism Cyberattacks on Critical Infrastructure 3.2.3 Terrorism Offense-Defense Balance 3.3 Loss of Control Loss of Control 3.3.1 Loss of Control Erosion of Control 3.3.2 Loss of Control Unleashed AI Agents 3.3.3 Intelligence Recursion Intelligence Recursion 3.4 Existing AI Strategies Existing AI Strategies 4 Deterrence with Mutual Assured AI Malfunction (MAIM) 4.1 MAIM Is the Default Regime 4.2 How to Maintain a MAIM Regime 5 Nonproliferation 5.1 Compute Security 5.1.1 Compute Security Export Controls 5.1.2 Compute Security Firmware-Level Features 5.2 Information Security 5.2.1 Information Security How to Improve Information Security 5.3 AI Security AI Security 5.3.1 AI Safeguards Malicious Use 5.3.2 AI Safeguards Loss of Control 6 Competitiveness 6.1 Military Strength 6.2 Economic Security 6.2.1 Economic Security Manufacture AI Chips 6.2.3 Economic Security Facilitate Immigration for AI Scientists 6.3 Legal Frameworks Governing AI Agents 6.3.1 Legal Frameworks Governing AI Agents Aligning Individual AI Agents 6.3.2 Legal Frameworks Governing AI Agents Aligning Collectives of AI Agents 6.4 Political Stability Political Stability 6.4.1 Political Stability Censorship and Inaccurate Information 6.4.2 Political Stability Automation 7 Conclusion Conclusion 8 Appendix: FAQs