Jacob Coxon's AI "kill us all" alarm doubles as a sales pitch

The piece places today's AI-safety alarms in a lineage. In March 2023, more than 1,000 intellectual luminaries and corporate oligarchs signed an open letter warning of "profound risks to society and humanity" and calling for a moratorium on more powerful AI systems; Elon Musk and Steve Wozniak were among the signatories. The commentary notes those concerns arrived hundreds of billions of investment dollars before the latest AI-extinction panic.

The argument adds that focusing on dramatic, cinematic extinction scenarios — HAL 9000 or Skynet — obscures the slower-moving environmental costs of AI in electricity, water, and other resources, and the cognitive attrition in education as students hand off thinking to AI. As the piece puts it, outsourcing thought to AI may leave people able to do less, a deskilling it treats as more fundamentally human than what earlier automation took.

It also ties the sales pitch to national security: even as Anthropic ostensibly pleads for guardrails, the argument goes, the United States will not let China get there first. That reasoning, the professor concludes, turns "if anyone builds it, everyone dies" into a pitch for exactly the action being cautioned against — a pattern the commentary compares to earlier technologies like electricity, the telegraph, the phone and radio, and to Meta's Metaverse writedown.