Policy makers can shift the focus by tying regulatory approvals to current socio-economic impact assessments. Instead of debating theoretical scenarios where machines might take over, lawmakers should demand rigorous audits of existing models. These audits must check for discriminatory patterns in hiring algorithms and track how automation affects local labor markets. We can also implement strict liability frameworks. If a company deploys a biased tool that denies someone a loan, they should face immediate legal consequences. This makes harm tangible and expensive, forcing companies to care about social outcomes today.
Another mechanism involves restructuring funding and research priorities. Currently, much of the high-level policy debate revolves around long-term safety. Governments should redirect a portion of these technical safety budgets toward labor retraining programs and community-led oversight boards. Giving affected workers a seat at the table changes the conversation. When people whose livelihoods are actually changing enter the room, the discussion moves away from sci-fi scenarios and back to real-world survival.
Finally, transparency mandates help. Requiring companies to publish data on their training sets and the wages of data annotators prevents them from hiding exploitation behind complex code. Transparency turns abstract worries into documented facts.
