The Ex-Congressman Who Says AI Isn't Unstoppable — Brad Carson

2026-05-31 · Source: Machine Learning Street Talk · Field: Government & Public Sector — Public Policy & Governance, Regulatory & Compliance, International Relations & Diplomacy · Depth: Advanced, extended

Summary

Brad Carson, a former Congressman, argues that AI's future is not predetermined and can be actively shaped through regulation, challenging the notion of an "unstoppable" technology. He highlights the US's control over critical AI chips as a key lever to influence global AI development. Carson emphasizes that AI should be treated as a product, not a human, rejecting claims of First Amendment rights for models and advocating for product liability in cases of misuse like deepfakes or suicide encouragement. He criticizes the opaqueness and probabilistic nature of AI in warfare, which complicates accountability, and dismisses the "inevitable AI arms race" as a dangerous myth, citing historical examples of successful technology regulation. Carson also stresses the importance of international dialogue, particularly with China, on AI governance and warns against the US's historical over-reliance on technical solutions in conflict, underscoring the enduring importance of the human element. He notes the low public trust in AI, which poses a significant threat to its long-term societal acceptance.

Key takeaway

For policy makers drafting AI legislation, recognize that AI is a controllable product, not an unstoppable force. You should prioritize establishing clear product liability frameworks for AI misuse and mandate engineering controls to prevent harmful outputs. Engage in international dialogues, especially with China, to foster global AI governance and avoid dangerous arms races. Your actions are crucial to build public trust and prevent a backlash against the AI sector.

Key insights

AI's future is governable, not inevitable, requiring active regulation, accountability, and international cooperation.

Principles

• AI should be treated as a product, not a person with rights.
• Technology regulation is achievable, countering fatalistic "inevitable" narratives.
• Over-reliance on technical solutions often betrays human problems.

Method

Mandatory testing and evaluation of frontier AI models should be implemented, potentially through independent verification organizations overseen by a regulatory body, to ensure public oversight without creating large bureaucracies.

In practice

• Apply product liability laws to AI for misuse like deepfakes.
• Engineer AI models to prevent harmful outputs, such as suicide encouragement.

Topics

• AI Governance
• AI Regulation
• Product Liability
• Autonomous Weapons Systems
• Chip Export Controls
• Public Trust in AI

Best for: Policy Maker, AI Ethicist, Legal Professional

Related on AIssential

• America, and probably the world, stands on a precipice. — Unrestricted military AI deployment and executive circumvention of Congress pose critical risks to global safety.
• Sustaining AI safety: Control-theoretic external impossibility, intrinsic necessity, and structural requirements — External control cannot sustain AI safety; intrinsic safety, based on internal objectives, is necessary.
• Links For April 2026 — Diverse topics reveal underlying principles in AI, social dynamics, and governance, often challenging conventional wisdom.
• I Robot. — Asimov's Laws of Robotics offer a foundational, albeit flawed, framework for AI safety and alignment discussions.
• Every Debate On Pausing AI — An international, bilateral AI pause is distinct from a unilateral one, requiring mutual agreement and enforcement.
• Cresti’s core claim is simple: if you can’t explain where data goes and who can access it, you cannot credibly claim governance. — AI's integration into devices necessitates robust traceability and control to maintain digital sovereignty and institutional competence.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine Learning Street Talk.