How can regulators design AI rules that mitigate societal harms without creating high compliance costs that favor massive tech corporations?

Regulatory capture happens when heavy compliance burdens act as a moat. Large firms can afford massive legal teams to navigate complex rules, while smaller startups buckle under the paperwork. To stop this, lawmakers should adopt a tiered approach. Instead of applying the same rules to every company, tie the level of oversight to a firm's actual market power and risk profile.

Focusing on specific use cases rather than broad technology categories helps too. Regulating a medical AI used in surgery requires intense scrutiny; regulating a basic chatbot for recipe suggestions does not. By targeting high-stakes applications, we protect people without strangling niche innovators.

Openness is another safeguard. Mandating interoperability and data portability ensures that users can switch platforms easily. This prevents