Goliath Super Intelligence
IndustryOctober 20, 20262 min read

Illinois Governor warns AI could extend lifespan or act like nuclear weapons

In a October 2026 interview, Gov. JB Pritzker linked AI’s promise of longer lives to existential threats, urging regulation, antitrust enforcement, and safeguards against job loss as he seeks re-election.

In a October 19, 2026 episode of The Verge’s Decoder podcast, host Nilay Patel sat down with Illinois Governor JB Pritzker as the governor campaigned for a second term. The conversation, recorded amid a heated midterm cycle, turned to the governor’s views on artificial intelligence, data-center incentives, privacy protections and the evolving relationship between government and big tech. Pritzker framed the discussion as a test of how policy must adapt to rapid technological change.

Pritzker warned that the same AI breakthroughs that could help people live to a hundred years old might also pose threats comparable to nuclear weapons. He said, “AI could help us live to 100, or threaten us like nukes,” underscoring a belief that the technology’s benefits and dangers are tightly coupled. The governor argued that without clear safeguards, the existential risk could outweigh the promise of longer, healthier lives.

The governor described AI as a classic collective-action dilemma that cannot be solved piecemeal by individual states. He argued that federal and state agencies must coordinate to set guardrails, invest in research, and protect vulnerable populations from rapid automation. Pritzker emphasized that “some problems are just collective action problems that require the government to do a lot of work,” suggesting a permanent expansion of public-sector responsibility in the digital age.

When asked about antitrust policy, Pritzker cited a recent social-media exchange in which David Sacks retweeted Lina Khan’s warning that AI firms should not receive special exemptions. He agreed that “these companies don’t need antitrust exemptions” and called for more competition in the sector. The governor’s stance aligns with a broader bipartisan push to prevent market concentration that could amplify both economic and existential risks associated with advanced AI systems.

Pritzker warned that AI-driven automation could push the United States’ unemployment rate well beyond the current 4 percent benchmark. He projected a possible rise to 7 or even 10 percent, describing such a shift as “the baseline” for future economies. The governor stressed the need for policies that preserve the “dignity of work” and provide sufficient income, arguing that without proactive measures, large segments of the workforce could be displaced by machines.

Beyond AI, Pritzker highlighted data-center incentives and privacy as flashpoints of a bipartisan backlash against unchecked tech expansion. He noted that states compete for data-center investments through tax breaks, a practice he believes should be balanced with robust privacy safeguards. Linking these issues to a “generational political realignment,” the governor said the Democratic Party is becoming more aggressive in defending taxpayers and vulnerable citizens from the excesses of powerful technology firms.

Sources

  1. JB Pritzker thinks AI could help us live to 100 — or threaten us like nukes The Verge

More reports

Industry · October 20, 2026 · 2 min

New York Mayor Calls for Local Pushback as Federal AI Regulation Lags

Mayor Zohran Mamdani announced a school AI ban and housing disclosure rules while urging residents to resist unchecked AI adoption amid federal regulatory silence.

Industry · October 20, 2026 · 1 min

Anthropic and Nscale negotiate second $45 billion AI compute agreement

According to Bloomberg, the two firms are close to finalising another $45 billion contract for AI compute capacity, even as both prepare for initial public offerings.

Industry · October 20, 2026 · 1 min

Hugging Face warns embedded AI evaluators face reproducibility hurdles

Chief policy officer Irene Solaiman said at TechCrunch Disrupt that benchmarking lacks standardization, reproducibility and clear comparison methods, making embedded evaluator ecosystems difficult to implement.