Sam Altman just published OpenAI's most revealing document yet: a 13-page policy blueprint asking Washington to tax AI profits, create a national wealth fund paying dividends to every American, and implement a 4-day workweek. The proposal explicitly states we're "beginning a transition toward superintelligence" and calls for "containment playbooks for rogue autonomous AI" that can't be shut off. Altman wants a sovereign-style fund seeded by AI companies, modeled after Alaska's oil dividend program.
This isn't typical Silicon Valley policy theater. The CEO of an $852 billion company doesn't ask the government to redistribute his industry's wealth unless he genuinely believes that industry will eliminate most jobs. Altman's timeline matters here—he's not talking about distant sci-fi scenarios but "urgent" preparations for changes already underway. The document reads like someone who's seen OpenAI's internal roadmaps and concluded that current economic structures won't survive contact with what's coming.
What's missing from Altman's pitch is any acknowledgment that his own company's secrecy makes these preparations nearly impossible. You can't plan for superintelligence when the people building it won't share timelines, capabilities, or safety measures. Meanwhile, The New Yorker's investigation into Altman, based on internal memos from former chief scientist Ilya Sutskever, suggests a pattern of deception that makes his public policy recommendations feel disconnected from OpenAI's actual governance.
For developers and AI builders, Altman's economic predictions carry more weight than his policy solutions. If OpenAI's CEO thinks robot taxes are necessary, start planning for a world where AI dramatically reduces human labor value—regardless of whether Congress acts on his recommendations.