How Can Society Safeguard Everyone in the Age of Super‑Intelligent AI?

OpenAI recently released a strategic document titled “Industrial Policy for the Intelligence Age,” which outlines a vision for managing the societal impact of increasingly powerful artificial intelligence. It argues that AI has moved beyond simple assistance and is heading toward super‑intelligence that could outpace human capabilities, demanding proactive policy responses.

Protecting the Average Person's Wallet

The paper proposes a universal public wealth fund that would turn AI‑generated economic growth into shared dividends for all citizens, regardless of financial literacy or capital. Funding could come from higher taxes on capital gains and profits generated by automation, as well as direct contributions from tech companies and governments. The goal is to prevent wealth concentration and ensure that efficiency gains translate into tangible employee benefits, such as increased retirement contributions, healthcare subsidies, and family care allowances.

To sustain social safety nets, the authors recommend shifting tax bases from labor to capital, implementing progressive capital‑gain taxes, and possibly taxing the use of automated labor itself. They also suggest a flexible, adaptive safety‑net system that can automatically expand with rapid cash assistance, wage insurance, and training vouchers when unemployment spikes or certain industry‑specific risk thresholds are breached.

Human Touch and Scientific Explosion

While machines take over repetitive tasks, the article emphasizes the growing value of jobs that require human empathy—childcare, eldercare, education, healthcare, and community services. It calls for government investment in training for these sectors, higher wages, and better working conditions to help displaced workers transition into these roles. Additionally, it proposes small‑grant programs and AI‑powered startup toolkits to enable ordinary workers to launch micro‑enterprises, leveraging AI to handle administrative burdens.

Putting a Bridle on Super‑Intelligence

The authors warn that powerful AI systems introduce new risks, including malicious use, unintended autonomous behavior, and the potential release of dangerous models. They advocate for rapid development of protective models, red‑team/blue‑team simulations, and a robust trust stack that provides verifiable signatures for AI‑generated content and actions. Mandatory audits for frontier models, strict compliance standards, and transparent reporting mechanisms are recommended to prevent misuse.

Finally, the paper calls for modernizing information‑sharing frameworks, enabling citizens and watchdog groups to audit government AI deployments, and fostering international cooperation to mitigate cross‑border AI threats. It concludes that no single policy tool can address the challenges of super‑intelligent AI, and a multi‑pronged, inclusive approach is essential.