There’s a specific, practiced kind of caution that Wall Street hates. It’s the corporate speak that translates, in the cold light of a trading screen, to uncertainty. And uncertainty, in a market already jittery about the astronomical costs of the artificial intelligence race, is a direct path to a sell-off. That’s the simple story of what happened to Meta Platforms after its latest earnings report. The tech behemoth didn’t miss by a mile on revenue; it actually beat. It didn’t slash guidance; the outlook was merely a whisper light. The real gut punch was delivered in a single, seemingly innocuous line from Chief Financial Officer Susan Li: “We aren’t providing a specific outlook for 2027 capex at this time.”
Let that sink in. In the high-stakes poker game of AI infrastructure, Meta just declined to show its cards for 2027. For investors already nursing losses from a nearly 30% year-to-date decline in the stock, this wasn’t caution. It was a red flag waved directly in front of the “overbuilding” narrative that has plagued the tech sector all summer. The immediate result was predictable: a 10% plunge in early trading, vaporizing tens of billions in market value in a matter of hours. This isn’t just a bad day; it’s a signal that Meta stock is likely to remain dead money for the foreseeable future, trapped under the cloud of its own ambitious, yet opaque, spending plans.
The irony is thick enough to cut with a knife. Meta’s actual quarterly performance had some bright spots. The company hauled in $60.8 billion in revenue, edging out the $60.2 billion analysts surveyed by LSEG were expecting. The user base across its family of apps remains a colossus. The problem was entirely on the bottom line. Earnings per share came in at $6.18, a stark miss against the $7.14 consensus. CFO Li was quick to point to two specific, non-operational culprits: a $2.4 billion charge for legal contingencies and a $1.2 billion severance charge. On a call covered by CNBC, she noted the company faces “scrutiny on youth-related issues in several markets,” with U.S. trials this year that “may ultimately result in a material loss.”
- Legal contingencies charge of $2.4 billion
- Severance charge of $1.2 billion
- Third-quarter revenue forecast of $61 billion to $64 billion
- Capital expenditure outlook for this year tightened
- Low-end of the range for 2024 raised to $135 billion
- High-end remains at $145 billion
These are real costs, but Wall Street often looks past one-time charges. What it can’t look past is the future. And that’s where the guidance landed with a thud. The third-quarter revenue forecast of $61 billion to $64 billion bracketed, but didn’t quite leap over, Wall Street’s $63.1 billion hope. More critically, the capital expenditure outlook for this year was tightened, but not in the way bulls wanted. The low end of the range for 2024 was raised to $135 billion from $125 billion, with the high end staying at $145 billion. This wasn’t a cut; it was a confirmation that spending remains on a skyward trajectory. Then came the kicker about 2027. In the world of infrastructure planning, where data centers are multi-year commitments, a refusal to project spending three years out is read as an admission of profound unpredictability. Li’s follow-up comment, that “infrastructure planning remains highly dynamic,” only cemented the unease. It told analysts that even the 2024 numbers aren’t set in stone.
The market’s interpretation was swift and severe. As KeyBanc analyst Justin Patterson put it in a research note covered by Bloomberg, “Meta alluded to medium-term capex growth and issued 3Q guidance that was roughly in line with the Street, which feeds into AI bear concerns on overbuilding and ROIC.” The acronym ROIC—return on invested capital—is the heart of the matter. Investors are no longer simply applauding any and all AI investment. They are demanding a visible, believable path to profitability on these expenditures. When a company as large as Meta declines to even sketch that path for 2027, it implies the spending could be open-ended, with the payoff horizon receding further into the distance.
This creates a fundamental problem for the stock’s narrative. For months, the bear case on Meta and other AI-heavy spenders like Microsoft and Google has been one of a capital destruction cycle: a brutal arms race where companies are forced to spend billions just to keep pace, potentially eroding their legendary profit margins for years. Meta’s earnings call provided the bears with fresh ammunition. By being coy, the company validated the fear rather than dispelling it. In the current environment, where every data point is magnified, choosing not to give a number is often interpreted as being afraid of what that number might be.
The contrast with the recent past is telling. There was a time, not long ago, when Meta’s relentless focus on efficiency and its “year of efficiency” mantra propelled the stock to new highs. CEO Mark Zuckerberg won back Wall Street’s favor by ruthlessly cutting costs and showing discipline. Now, the script has flipped. The market is questioning whether the AI pivot, for all its long-term promise, is a return to the free-spending, metaverse-era Zuckerberg that made investors so nervous just a couple of years ago. The lack of a 2027 capex guide feels, to many, like a return to that less predictable era.
So, where does this leave shareholders? In a holding pattern, most likely. The combination of a bottom-line miss, a lightish guide, and a profound lack of clarity on the single most important cost driver for the next decade is a toxic mix. It will take more than one quarter of solid execution to dig out of this hole. The company needs to start providing more tangible evidence that its AI investments are generating incremental revenue and profit, not just consuming capital. It needs to articulate a clearer roadmap.
| Metric | Actual | Analysts Expected |
|---|---|---|
| Revenue | $60.8 billion | $60.2 billion |
| Earnings per Share | $6.18 | $7.14 |
| 3Q Revenue Guidance | $61 billion – $64 billion | $63.1 billion |
| 2024 Capex Low End | $135 billion | $125 billion |
| 2024 Capex High End | $145 billion | N/A |
| 2027 Capex Outlook | Not Provided | N/A |
Until it does, the specter of “dead money” will linger. The stock may bounce on oversold conditions or a broader market rally, but a sustained re-rating higher seems off the table until Meta can confidently tell the market what 2027 is supposed to look like. For now, the only thing that’s clear is that the market doesn’t like the silence.
