Tech giants, excluding Apple, are no longer asset-light. Their massive AI infrastructure investments mean they must now be judged by traditional metrics like Return on Invested Capital (ROIC), a standard previously applied to industrial companies, not software firms.
To achieve a reasonable return on investment, the current capital expenditure in AI by hyperscalers requires generating $2.5 trillion in new revenue. This amount exceeds the total current revenue of all Big Tech companies combined, highlighting a potential valuation bubble.
NVIDIA is financing its customers to buy its own chips, a move that could be seen as artificially inflating demand. While common in CapEx-heavy industries, the unprecedented scale raises questions about whether NVIDIA is propping up a bubble by acting as both supplier and financier.
A labor economist argues the focus on AI's potential job displacement distracts from a more immediate crisis: the US's poorly designed unemployment system. Historical tech adoption is slow, and improving tangible worker support policies should be the priority over speculative AI debates.
Labor economist Catherine Ann Edwards critiques UBI not just practically, but for its underlying tone. She argues it's a dismissive posture that gives up on workers, whereas a robust unemployment system should actively help them retrain, relocate, and re-engage with the workforce.
Allies like Europe and Latin America no longer trust the US as a stable partner, forcing them to build their own defense and trade systems independent of America. This loss of trust creates a long-term economic threat that a single 'reasonable' election cannot easily reverse.