A key distinction in AI regulation is to focus on making specific harmful applications illegal—like theft or violence—rather than restricting the underlying mathematical models. This approach punishes bad actors without stifling core innovation and ceding technological leadership to other nations.

The emergence of powerful, uncensored open-weight models like Obliteration.ai's demonstrates that safety guardrails from companies like OpenAI are easily bypassed. This suggests the long-term solution for AI safety won't be technical restrictions at the model level, but rather legal and regulatory enforcement.

If an AI model can identify that a user is planning a violent act, the operating company should be legally required to notify authorities. This parallels existing liability laws for professionals like bartenders who observe imminent danger, applying a "duty to report" standard to AI platforms.

When an AI agent errs in a medical or financial context, it is legally unclear who is liable: the AI lab, the deploying company, or the end-user. This novel legal problem, which challenges a century of precedent, creates significant friction and will slow agent adoption in regulated industries.

Instead of trying to legally define and ban 'superintelligence,' a more practical approach is to prohibit specific, catastrophic outcomes like overthrowing the government. This shifts the burden of proof to AI developers, forcing them to demonstrate their systems cannot cause these predefined harms, sidestepping definitional debates.

Demis Hassabis argues that market forces will drive AI safety. As enterprises adopt AI agents, their demand for reliability and safety guardrails will commercially penalize 'cowboy operations' that cannot guarantee responsible behavior. This will naturally favor more thoughtful and rigorous AI labs.

Legal systems are built around human accountability. When a Frontier AI independently launches attacks, governments face a crisis: who is responsible? The AI's owner, its user, or the AI itself? This lack of precedent for a non-human criminal paralyzes the development of effective regulation.

An FDA-style regulatory model would force AI companies to make a quantitative safety case for their models before deployment. This shifts the burden of proof from regulators to creators, creating powerful financial incentives for labs to invest heavily in safety research, much like pharmaceutical companies invest in clinical trials.

The U.S. has a built-in mechanism for AI safety that precedes formal regulation: the court system. The potential for lawsuits (tort law) incentivizes model makers to act responsibly, acting as a form of self-regulation that doesn't require a slow-moving government bureaucracy.