Meta's 1.9% Rebound Shows AI Confidence Is Back-but Only for Spend With Proof

Meta's rebound recovered only part of its loss

Meta's rebound was real, but it was only a partial repricing. After the stock was down 8%, the approximately 1.9% rebound looked less like full confidence restored and more like investors regaining composure after a selloff driven by higher capital expenditure and weaker free cash flow. In other words, the market was not rejecting AI outright. It was demanding clearer economic returns on AI spend.

The underlying business still looked healthy. Meta reported about $60.8 billion in second-quarter revenue, up 28%, and earned $6.18 a share before legal and severance pressures. Free cash flow, however, fell to $784 million from $8.55 billion a year earlier, and the company narrowed its 2026 capital-expenditure forecast to $130 billion to $145 billion. That is why the debate shifted from pure growth enthusiasm to the cash cost of building for AI. Bulls can still point to a large opportunity. Bears can fairly insist that investors want faster proof of returns.

Amazon reset the market standard for AI spending

Amazon did not bring back indiscriminate AI enthusiasm. It restored confidence that heavy AI spending can still be economically justified. The clearest signal was that demand was feeding through quickly enough to matter: AWS posted its fastest growth in 18 quarters, while operating income rose 43%. That helps explain the shift in tone from simple fear of capex to a higher burden of proof.

Why investor psychology changed

The recent market behavior was understandable, if a bit driven by recent pain. After the previous week's selloff, investors focused hard on the starkest warning sign they had seen: heavy spending without an obvious payoff became difficult to please. When returns are uncertain, markets tend to punish cash outflows more harshly and chase clearer evidence instead. That mood had already weighed on shares before Amazon reported.

Amazon reduced that uncertainty. Management gave investors a clearer link between capital spending and earnings, showing how AI investment is expected to generate returns. The message was not that AI spending is automatically forgiven. It was that investors now want to see who is paying and where the margin comes from.

What Amazon proved, and what Meta still has to prove

For Meta, the issue is not whether AI matters. It is that investors are now applying the same proof standard Amazon helped set. After being down 8% on weak free-cash-flow headlines, Meta's later rebound suggests panic eased, not that the market has fully accepted the company's spending profile.

What Meta likely still needs to show: - that advertising, business agents, and other products can convert AI investment into revenue faster than feared, - that free cash flow can recover from the current squeeze, and - that its narrowed 2026 capex range reflects demand and discipline rather than just a longer payback period.

That is why Amazon's report mattered for Meta investors. The bar did not move toward blind AI optimism. It moved toward AI spend with proof.