Ask HN: How do tariffs interplay with the AI Bubble?
Many “AI” tools either don’t work or don’t work reliably & consistently enough to be used in production.
For a while now things tech felt increasingly like is on a path for reckoning similar to the dot com bubble, where company valuations were headed for a major correction with a long recovery even if the underlying technology was going to upend life in time.
I’m having trouble figuring out how the impact of tariffs changes this course. Curious what others think.
Stock market valuations are experiencing a correction, but unlike dotcom few “AI era startups” have gone public. The job market in tech is still out of whack from 2022 and feels like a tale of have and have nots (those employed with the right skills can’t keep up, but those unemployed are struggling to get through the noise and find a job).
I see a world in which tariff nonsense corrects for valuations and prevents a deeper correction in tech, see a world where one hits the other back to back to cause serious problems, and see a world where I’m missing the forest for the trees.
The AI trade bubble has already been popped, early by unrelated macro events.
Big companies are still investing in centers, and nvidia is still supply constrained to my knowledge however. New models are popping up. So there’s still real business/progress happening.
Remains to be seen how useful the products will ultimately be. Bottom line, growth should be smaller and more realistic from here on. Though I’m sure there’ll be spikes when the speculator contingent comes back to the market.
The later that happens, the more they’ll be constrained by reality.
The AI/semiconductor trade has popped, but I don’t think tech companies basing their business model on it has.
What I think you are suggesting is that it’s possible for companies to move on from AI hype without a dotcom-level correction. More akin to the way the metaverse and NFTs did. And maybe you aren’t wrong.
I was a pre-teen in 2000, so while I remember the hype, froth and general ridiculousness of the late 1990s (was just explaining Priceline for groceries to someone yesterday), I don’t remember the particulars of how big it got or quickly it ended.
As a product with use cases AI seems more comparable to the internet than the metaverse and NFTs though.
AI-only companies like OpenAI might flame out, but I think BigTech won’t be materially affected either way. They are just too big. Losing a few years petty cash on the bet will maybe affect executive bonuses but won’t hurt much longer term.
So you think that pre-AI (say early 2022), there was no startup bubble? And valuations we had back then, were sane?
No, but I think we saw the start of a correction for that in 2022 and 2023. One which the emergence of AI sort of interrupted in a way that feels muddy.
A lot of startups still are trying to avoid raising down-rounds because of valuation used to raise money in 2021 that were unrealistic. For startups that seems like it is still one of the major issues that will come to a head. Particularly as it becomes clear AI is not yet mature enfough to empower the kind of growth without headcount people are projecting for.
The 11X story last month seemed like a leading indicator we were starting to see AI’s real vs touted ability come to a head, but not sure where that goes in light of tariffs.