Instead of trying to anticipate every potential harm, AI regulation should mandate open, internationally consistent audit trails, similar to financial transaction logs. This shifts the focus from pre-approval to post-hoc accountability, allowing regulators and the public to address harms as they emerge.

A rapid, significant (e.g., 5%) spike in unemployment over a short period (e.g., 6 months) due to AI would trigger an immediate and massive political and economic response. This would be comparable in speed and scale to the multi-trillion dollar stimulus packages passed during the COVID-19 pandemic.

AR Rahman believes AI tools that can replace human jobs are a destructive force that must be regulated. He compares it to firearms, arguing that just as there are rules for ownership, there should be rules preventing the deployment of AI that makes entire skill sets worthless.

When addressing AI's 'black box' problem, lawmaker Alex Boris suggests regulators should bypass the philosophical debate over a model's 'intent.' The focus should be on its observable impact. By setting up tests in controlled environments—like telling an AI it will be shut down—you can discover and mitigate dangerous emergent behaviors before release.

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.

Technical research is vital for governance because it provides concrete artifacts for policymakers. Demonstrations and evaluations showing dangerous AI behaviors make abstract risks tangible, giving policymakers a clear target for regulation, aligning with advice from figures like Jake Sullivan.

The potential rise in unemployment from AI will not happen in a vacuum. Central banks and governments are expected to use tools like interest rate cuts, unemployment benefits, and targeted spending to stimulate the economy, thereby shortening and reducing the severity of any labor disruption.

Recent AI discourse shows a shift in petition effectiveness. Vague, doomsday-focused letters had little impact. In contrast, a new Stanford-led petition on specific economic impacts is gaining resonance because it's grounded in current reality and less prescriptive, attracting actual industry builders.

Overly-specific regulation focused on AI tools (e.g., model size) risks accidentally stifling valuable, unforeseen use cases. A better policy focuses on outcomes. For example, prosecute fraud committed with an LLM, but don't regulate the LLM itself, thereby protecting innovation while punishing misuse.