Banks Just Found Out What AI Debt Is Virtually Worth: Zero

YouTube · Watch/Read · (gathered 2026-10-18)

• El argues that while AI technology is spectacular and has real potential, its current stewardship is a “disaster,” leading to a disconnect between its astronomical valuations (e.g., OpenAI’s $1.5 trillion ask) and its actual delivery and precarious financial structure.
• Public trust is eroding due to misuse (e.g., Flock Safety’s surveillance cameras used for stalking, leading to staff hemorrhaging and contract cancellations) and employee exodus over safety concerns.
• While AI shows 15-50% productivity gains in some tasks, 95% of enterprise generative AI pilots fail to deliver measurable business impact, and only 20% of organizations report revenue growth, echoing the “Electric Dynamo” analogy where real gains take decades and fundamental organizational change.
• The AI industry is built on a layered financial structure, with companies like OpenAI projecting massive compute spending ($856 billion by 2030) funded by debt, often guaranteed by companies like Nvidia, creating a system where the risk is transferred and the entire structure depends on future revenue materializing at an enormous scale.
• The market is already showing signs of strain, with AI-linked debt (e.g., Project Jupiter loans) repricing at a discount (89-91 cents on the dollar) and corporate bonds trading at wider spreads, indicating lenders are demanding better terms as capital expenditure outpaces monetization, creating “localized bubble dynamics” within a real technological revolution.