332. That is how many views the Straits Times business desk post had when it relayed Jensen Huang’s latest message to a jittery AI market. A YouTube breakdown from a month earlier, titled “Nvidia CEO Jensen Huang Makes Shocking Prediction About AI Market,” had 25,466 views. Both are tiny numbers next to the amount of money riding on whether Huang is right.
I review AI toolkits for a living. My job is to figure out what works and what doesn’t, so I’m not the person to tell you where Nvidia stock goes next. I do have a view on what this kind of reassurance means for people building with these tools every day.
What Huang actually said
Speaking at a Bloomberg event in Berlin on October 16, 2026, Huang addressed the latest AI stock-market rout. Bloomberg described his outlook as sanguine. His line was short: “All of this is going to pass.”
He framed the volatility as temporary. He argued that AI’s impact on industries around the world will soon outweigh investor worries, pointed to long-term benefits for every industry and every country, and said we are only at the beginning of this shift.
None of that is new for Huang. Fortune has described him as the biggest foil to AI doomerism about existential risk to humanity. He is the optimist in the room, consistently. Berlin was the same message aimed at a different worry. This time the fear isn’t that AI ends the world. It’s that AI ends a lot of portfolios.
Consider who is talking
Nvidia sells the hardware much of the AI industry runs on. When the CEO of that company says a downturn will blow over, he is also describing his own business outlook. That doesn’t make him wrong. It makes him an interested party, and his words should be read that way.
There’s an old line about gold rushes: the people who reliably made money sold the shovels. Huang sells the shovels. His confidence that people will keep digging is sincere, and it’s also exactly what you’d expect from him.
My reviews follow the same rule. If a vendor tells me their product is great, I thank them and then run my own tests. Huang’s statement is a vendor statement. It’s useful information, but it isn’t a benchmark.
Why the market mood matters less than you think
The stock market and the toolkit market run on different clocks. Investors price expectations months or years out. Developers care whether a tool cuts their build time this sprint.
Look at what shipped just this summer. On July 10, 2026, GPT-5.6 launched, and SpaceXAI released Grok 4.5, a model that bets on speed and cost over raw power. That second release says more about where things are heading than any stock chart. When a major lab competes on price and latency instead of top-line capability, the tools are maturing into products people pay for and use, not just demos built to impress investors.
Whether or not “all of this” passes on Wall Street, those models exist, they’re in developers’ hands, and they either solve problems or they don’t. A sell-off doesn’t make a working tool stop working. A rally doesn’t fix a broken one.
What I’d tell a team picking AI tools right now
If market turbulence is making your leadership nervous about AI spending, a few practical points:
- Judge tools by what they do in your workflow, not by the share price of the company behind them.
- Favor tools that let you switch models without rewriting your stack. If one vendor stumbles, you want an exit.
- Watch cost per useful output, not headline capability. The industry’s own push toward speed and cost suggests buyers care about this.
- Be wary of anyone, CEO or influencer, whose case for a tool is mostly about the future.
That last point cuts both ways. Huang’s “we are just at the beginning” pitch is about the future. So is every doom-laden prediction that the bubble is about to burst. Neither tells you whether the code assistant you’re trialing writes code that passes review.
My take
I think Huang is probably right about the long arc and less useful about the short one. AI tools are clearly earning a place in real work. I see it in my testing every week. The useful ones keep getting cheaper and faster, and the gap between “impressive demo” and “thing I’d put in production” keeps shrinking.
“Going to pass” is not a timeline, though. It doesn’t tell a startup founder how long to stretch their runway or tell a CTO whether to sign a three-year contract. Optimism is free when you sit at the top of the supply chain. Everyone downstream has to make decisions with real budgets and real deadlines.
So treat the Berlin comments as what they are: a confident forecast from someone with excellent visibility and an obvious stake. Then go back to evaluating tools the boring way. Run them on your data. Measure what they save you. Cancel the ones that don’t earn their seat.
Markets will swing. That 332-view post will scroll past and be forgotten. The tools that work will still be working next quarter, and that’s where I’ll keep my attention.
🕒 Published: