Big Tech uses guarantees to keep $300B AI exposure off balance sheets

andsoitis

The headline invites the reader to think these liabilities are concealed from investors, but they're not. Deeper in the article it talks about how credit-rating agencies explicitly includes in their modeling. The guarantee allows lenders to behave partly as though they're lending to NVIDIA, Meta or Broadcom rather than to an AI startup or single-purpose data-center company. The interesting question isn't "Is the debt hidden?" (it isn't), but is who ultimately owns the downside risk of the AI infra boom? More interesting headline would be "Big Tech is using its creditworthiness to finance the AI boom without taking on equivalent amounts of direct debt."

aurareturn

Why are people such big opponents of the AI buildout? I generally don't see many cons to it. 1. Big tech was just using profits to buy back shares before AI build out. Would you rather they buy back shares with their profits or finance the AI build out? 2. If there is an AI bubble, great, when it pops, we'll all have ultra cheap tokens and we'll have all these amazing AI infrastructure built for the next wave of innovation. 3. Even if AI progress stops here and models hit a wall in intelligence, we'll still have AI good enough to solve millennium problems. My theories on why people are so anti-AI: 1. It threatens their jobs. 2. They want AI stocks to drop so they can buy the dip. 3. Their identity is being challenged because AI makes what they do easier.

peri-cl

https://archive.is/a4uct