To: Johnny Canuck who wrote (70643 ) 2/10/2026 12:24:46 AM From: Johnny Canuck Read Replies (1) | Respond to "AI Data Centers Face Multi State Bans: Here's What's Happening," examines the massive financial and structural risks currently facing the AI industry. Miller argues that while tech giants are spending hundreds of billions on "token factories," the underlying math, legal protections, and physical infrastructure are beginning to crumble.The $650 Billion "Token Factory" Gamble Massive Spending: Four major companies (Meta, Amazon, Google, and Microsoft) plan to spend $650 billion this year alone on AI infrastructure [ 00:00 ].The Token Economy: Microsoft’s Satya Nadella describes data centers as "token factories," where the new measure of economic output is "tokens per dollar per watt" [ 03:02 ].Market Panic: Despite this spending, these companies recently lost $950 billion in market value as investors grew skeptical of the returns and the "circular financing" where big tech funds AI startups that then pay for their own cloud services [ 07:17 ].Questionable Accounting & "Creative" Math Michael Burry’s Warning: The investor who predicted the 2008 housing crash alleges that companies like Oracle and Meta are overstating earnings by 20–27% through creative depreciation schedules for AI assets [ 10:45 ].Subsidized Growth: Miller notes that OpenAI’s $20/month subscription is heavily subsidized; without Microsoft’s backing, it would likely cost $80/month [ 09:44 ].The Insurance Industry Pulls Out Uninsurable Risk: In a major red flag, the insurance industry is refusing to cover AI liabilities. AIG, Great American, and WR Berkeley have introduced absolute AI exclusions [ 18:54 ].The "Black Box" Problem: Insurers claim LLMs are too much of a "black box" to underwrite, meaning companies deploying AI may have no coverage if the technology fails or causes widespread damage [ 21:11 ].Political and Social Backlash State Bans: Six states, including New York, are considering moratoriums on new data centers due to concerns over energy costs and grid stability [ 16:42 ].Unusual Alliances: Political opposites like Bernie Sanders and Ron DeSantis have both expressed concerns over the impact of AI data centers on utility bills and society [ 16:58 ].Mass Layoffs: Companies are cutting tens of thousands of jobs (Amazon, Oracle, etc.) to free up capital for AI, even as AI researchers admit these models often cannot learn from their own mistakes [ 12:20 ].Physical Bottlenecks The "Dirt" Problem: Sam Altman recently confirmed that the primary bottleneck for AI is chips and physical materials (quartz/silicon) , not software [ 15:15 ].Failed Tech: Researchers at Apple and Yale have found evidence of "reasoning collapse" when AI models face problems outside their training patterns, questioning the path to AGI [ 15:02 ].Summary: Miller concludes that the AI boom is increasingly looking like a bubble built on "math that doesn't add up," facing a "red flag factory" of uninsurable risks, political revolts, and overstated earnings [ 23:52 ].
