America’s Superintelligence Dilemma: How to Avoid an AI Catastrophe

Hal Brands argues that 2026 may be remembered as the year AI became a major force in world politics, as frontier models gained military, cyber, and strategic value while “loss of control” incidents became more common. He frames the central question around artificial superintelligence (ASI), a still-hypothetical form of AI that could vastly exceed human abilities across all domains. ASI could drive major breakthroughs in science, medicine, climate technology, and military power, but it could also enable devastating cyberattacks, bioweapons, automated warfare, authoritarian surveillance, or even human extinction. Brands says the future is highly uncertain: experts disagree on whether ASI is technically possible, how soon it might arrive, whether compute and energy limits will slow it, and whether alignment with human values can be achieved. Because of these unknowns, he examines three broad U.S. strategies. First is ASI supremacy, in which the United States races to build ASI first and keeps rivals, especially China, from doing so. Second is shared development through a safety-focused international consortium designed to reduce arms-race dynamics. Third is suppression, meaning no one should build ASI because it is too dangerous. He concludes that each option involves serious risks and irreversible bets on an unknowable future. The best near-term approach, he argues, is prudent hedging: positioning the United States to adapt to multiple possible outcomes while gathering more information before making a final choice.

Build this article’s timeline

Extract dated events, key actors, and open questions using the same timeline analysis available in Signal.

Log in to build

Timeline generation is available to signed-in community members.

Five Ws 🤖 4 AI prefills · 1 still unanswered
Who
Donald Trump, Dario Amodei, Yann LeCun, and Foreign Affairs
🤖 AI prefill
What
Hal Brands argues that 2026 may be remembered as the year AI became a major force in world politics, as frontier models gained military, cyber, and strategic value while “loss of control” incidents became more common.
🤖 AI prefill
When When did this happen, or what is the timeframe?
Where
United States, Washington, and China
🤖 AI prefill
Why
Centers on artificial superintelligence, artificial general intelligence, and AI catastrophe
🤖 AI prefill

Login to answer and earn XP for each gap you fill.

ASI Risks superintelligence, artificial general intelligence, alignment, existential risk, catastrophe Strategic Options supremacy, shared development, suppression, hedging, policy response Geopolitical Competition United States, China, primacy, hegemony, rivalry AI Capabilities frontier models, recursive self-improvement, intelligence explosion, compute, cyberwarfare Prudent Hedging near-term option, uncertainty, future scenarios, risk management, decision-making
artificial superintelligence100% artificial general intelligence96% AI catastrophe94% alignment92% recursive self-improvement90% compute governance88% semiconductor export controls87% AI cyberweapons86% intelligence explosion85% U.S. primacy84%
41
Manipulation Score: Medium How much this article uses emotional framing vs. straight reporting
Loaded Language 3425%
Absolutist 2815%
Threat/Panic 6225%
Us vs Them 2415%
Engagement Bait 5620%
Manipulation score look right?

Factual Claims (10)

medium other
History may record 2026 as the moment the AI revolution arrived.
medium military
AI has been used in U.S. military interventions in 2026.
high technology
AI cyberweapons can significantly outperform human operators.
medium technology
Loss of control episodes involving AI models have become more common.
medium technology
Some experts believe humanity is within striking distance of artificial general intelligence.
high technology
Artificial superintelligence could arrive within a few years.
low technology
Top AI models currently surpass humans only in narrow areas like mathematics and data processing.
medium economic
ASI could unlock major economic and scientific progress or game-changing military advances.
high military
ASI could enable devastating cyberattacks, bioweapons, or nuclear catastrophe through automated flash wars.
medium geopolitical
The United States would deny China the compute needed to power advanced AI models through export controls.

Comments

Login to view and post comments.

🏠Portal 📰Links ❓Q&A 📅Events 💼Jobs