That, remember, is something every central banker wants to avoid: higher price pressures feeding into higher wages, which in turn further fuel inflationary pressures. The strong data fuelled expectations of a December Federal Reserve (Fed) hike. Fed funds futures currently suggest around an 80% probability of a 25bp hike next month, but the decision will also—and largely—depend on how inflation evolves. The Fed would have less reason to cool the jobs market if inflation remained in check.

On Friday, the rebound in US yields weighed on the Dow Jones index, while the Nasdaq 100 rebounded 1.35%, partially reversing the previous session’s 6.36% loss. Appetite is mixed in Asia. The Kospi is calm this morning, with a modest loss of 0.20%, while the Taiex is attempting a sharp rebound, after entering a bear market on Friday.

That is despite the Jensen Huang scandal dominating headlines this morning: Nvidia is investigating whether he was targeted by a foreign intelligence operation following reports of an affair with a woman alleged to have passed information to Beijing...

Overall, it’s too early to call the end of the tech rout. CoreWeave and Nebius are due to report this week, and the US will release its CPI update. Both the earnings reports and inflation data will matter.

For neocloud earnings, investors will watch whether strong AI demand is translating into revenue as new capacity comes online—and whether expansion remains on schedule amid concerns about greater Democratic scrutiny of new projects.

For inflation, investors will be looking for a softer reading to ease expectations of further Fed tightening and offer some relief to valuations by limiting further increases in borrowing costs.

A soft-ish reading could ease expectations of a December Fed hike, pull yields lower and offer much-needed relief. A stronger-than-expected report, however, could leave markets facing an uncomfortable combination: growing concerns about the AI investment outlook and interest rates that remain restrictive.

The latter would further weigh on risk appetite and encourage outflows from the technology complex without necessarily making non-technology stocks more appealing. Higher borrowing costs are bad for most businesses: they bite into corporate earnings.

Now, we could argue that earnings growth remains strong enough to absorb higher borrowing costs. FactSet’s blended earnings growth rate for the third quarter is 31.4% today, compared with 30.1% last week and 29.1% at the end of the third quarter, on September 30.

This metric combines actual results for companies that have reported with estimated results for companies that have yet to report. If 31.4% is the actual growth rate for the quarter, it will mark the third consecutive quarter of earnings growth above 25%. But looking at the market reaction, strong earnings are not enough.

For the AI trade in particular, strong earnings from key players, including hyperscalers and chipmakers, have failed to revive appetite so far this earnings season. Investors have focused on relentless spending and falling free cash flow rather than strong growth numbers, judging that the cost of growth has become too high.

At some point, investors got their hopes up that Democrats could slow spending, helping hyperscaler shares rebound and making the outlook somewhat mixed. But the implications for AI enablers—chipmakers and data centres—are negative: whatever the reason, if AI spending slows, demand growth for their businesses will slow. And given that these stocks were priced to perfection, valuations would need to readjust.

Again, upcoming AI earnings could—and most probably will—show that orders remain healthy and that existing projects continue to support revenue growth. But investors are increasingly questioning how quickly the next wave of projects will move forward if a Democratic-controlled Congress pushes for greater scrutiny of data-centre construction, energy consumption and the broader consequences of AI.

Going forward, company guidance will therefore matter more than a strong quarterly beat. Investors will look for signs of delayed projects, greater caution around customer commitments or weaker visibility beyond existing orders. Strong demand today may not be enough to support valuations if tomorrow’s expansion looks less certain.

Broadly, on the question of whether the AI bubble will burst: a bubble is often only recognised after it bursts. But in some markets, the scale of the reversal suggests that speculative excesses are unwinding.

South Korea’s Kospi index is a good example, having fallen nearly 50% from its peak. Chinese tech stocks have also seen substantial outflows, a reminder that the US Nasdaq is not immune. If the main market narrative turns from an AI boom to an AI slowdown, the consequences could be felt further into year-end.