OpenAI's Intelligence Age Policy Report Gets a Critical Read

In episode 453 of This Machine Kills, hosts Jathan Sadowski and Edward Ongweso Jr. do a close read of OpenAI's "Industrial Policy for the Intelligence Age: Ideas to Keep People First," a document the company released in April 2026 as a framework for surviving "the transition toward superintelligence." The hosts describe finding "some moments of insidiously clever thinking" alongside "a laundry list of boring ideas that repackage existing things as exciting innovations."

The report's proposals are ambitious on paper: treating AI access as a fundamental right, a four-day work week, a public wealth fund, tax reform for the AI economy, and giving workers a voice in how AI is deployed. OpenAI also offered up to $100,000 in grants and up to $1 million in API credits for research projects that advance these proposals. A critical analysis at TechPolicy.Press labeled the document a "policymercial," framing it as commercial marketing dressed up as substantive policy.

The sharpest criticism in that analysis concerns the gap between OpenAI's public advocacy and its track record. OpenAI opposed California's SB1047, which included provisions for third-party audits, incident reporting, safety protocols, whistleblower protections, and a public computing cluster called CalCompute. The industrial policy document now proposes similar measures. Meanwhile, OpenAI's internal superalignment research team received only 1-2% of its promised compute, and key figures including Jan Leike and Ilya Sutskever departed. TechPolicy.Press noted that "we still don't have solid data on how LLM productivity gains manifest or are distributed," a gap that makes the report's abundance promises hard to evaluate.

The honest read is that communities absorbing the direct effects of AI infrastructure expansion are expected to wait for promised benefits while governments subsidize that expansion. Whether any of the more genuinely redistributive proposals, like the public wealth fund or worker voice mechanisms, get taken up by policymakers independently of OpenAI's lobbying interests remains the open question this episode leaves on the table.