
Meta's stock fell 11% after reporting that while revenue grew 28% to $61bn in the second quarter, profits dropped 14% and the company raised its AI spending forecast to $130bn–$145bn annually. Investors are frustrated because Meta's ambitious AI investments—which consumed nearly all of the quarter's free cash flow—have not yet produced commercial products or revenue. CEO Mark Zuckerberg acknowledged the bet is large but said the company plans to sell AI models and services to other businesses and build autonomous AI agents; however, those revenue streams do not yet exist.
Summaries like this, in your inbox every morning.
Sign up free →What happened
Meta's stock fell 11% after the company reported Q2 revenue of $61bn (up 28% year-over-year) but profits fell 14% to $6bn. The firm raised its annual AI spending forecast to $130bn–$145bn, up from $125bn announced just three months earlier. Free cash flow for the quarter hit $784m, its lowest level in at least five years.
Why it matters
Investors are concerned that Meta's AI investments have not yet yielded commercial products or revenue streams. CEO Mark Zuckerberg said the company plans to sell AI technology to other businesses by 2028, and CFO Susan Li told analysts this would drive returns on spending—but such business lines have not yet materialized. One analyst compared the pattern to Meta's previous metaverse spending, which consumed tens of billions without user adoption.
What to watch
Meta is developing AI agents (autonomous chatbots) that Zuckerberg said will be the company's "next wave of product" in coming months and years, and plans to make its Muse Spark AI model easier for companies to integrate. Microsoft, by contrast, reported a 5% share rise on the same day despite $175bn in planned annual capital spending on AI infrastructure—suggesting investors reward companies when AI spending produces measurable financial returns.
Meta shares plunged 11% on Wednesday after the company reported second-quarter results that exposed a widening gap between revenue growth and profitability. Revenue for the April-to-June period grew 28% year-over-year to $61bn, but profits fell 14% to $6bn. The decline in profitability was driven by a dramatic acceleration in AI spending: Meta raised its full-year AI spending forecast to $130bn–$145bn, up sharply from $125bn the company had projected just three months earlier.
The most telling metric was free cash flow—the money Meta retained after paying for operations—which fell to $784m for the quarter, the lowest level the company has recorded in at least five years according to its financial records. One analyst at Forrester observed that "what it generated in cash this quarter almost all got eaten by AI infrastructure spending," highlighting the tension between Meta's cash-generation engine and its capital-intensive AI ambitions.
CEO Mark Zuckerberg defended the spending as necessary to accelerate the company's core business and acknowledged the scale of the bet. "I get that this is a big bet across the industry," he said. "My personal bet is that the people who invest in this will feel very good and be rewarded over time." He pointed to AI's role in driving engagement on Instagram and Facebook, as well as boosting smaller businesses' advertising capabilities. Zuckerberg also outlined two major future initiatives: AI agents—autonomous chatbots that he said will become the company's "next wave of product" in the coming months and years—and B2B sales of AI technology to other companies.
For the B2B effort, Zuckerberg said Meta would start by making its Muse Spark AI model easier for companies to integrate. "We expect to build a large business for large businesses," he said, adding that the opportunity extended beyond compute sales to API services and productivity tools. CFO Susan Li told financial analysts that selling this technology to other companies would help drive returns on AI spending, and suggested such revenue would materialize by 2028. However, these revenue streams do not yet exist, and investors expressed skepticism about whether Meta's growing list of AI initiatives represented diversification or distraction.
The market's reaction stood in sharp contrast to Microsoft's performance the same day. Microsoft reported Q2 sales up 18% to $90bn and profits up 31% to $36bn, and its stock rose 5% in after-hours trading. Microsoft also disclosed plans to spend $175bn on capital investments in the next year, mostly for AI and AI infrastructure. The key difference was that Microsoft's AI investments were visibly producing measurable returns, whereas Meta's had not yet yielded commercial products or revenue.
Meta's spending trajectory reflects a broader shift in how technology companies are approaching artificial intelligence, but Wall Street is drawing a sharp distinction based on financial returns. The company's Q2 results reveal a tension: revenue growth of 28% year-over-year to $61bn looks healthy on the surface, yet profits contracted 14% to $6bn and free cash flow plummeted to $784m—the lowest quarterly figure the company has disclosed in at least five years. The reason is straightforward: Meta's AI infrastructure costs consumed nearly all available cash.
What distinguishes Meta's position from competitors is the lack of demonstrated revenue from its AI bets. CEO Zuckerberg framed the spending as an investment in future capabilities—AI agents that will operate autonomously, and B2B sales of AI models and services to other companies. CFO Susan Li suggested such revenue would materialize by 2028. Yet as one analyst noted, these business lines do not exist yet, and the comparison to Meta's metaverse spending (tens of billions with minimal user adoption) carries real weight among skeptical investors. In contrast, Microsoft reported a 5% stock gain the same day despite announcing $175bn in annual capital spending on AI infrastructure, but Microsoft's AI investments are visibly tied to revenue growth: profits rose 31% to $36bn and sales grew 18% to $90bn.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
No comments yet. Be the first to share your thoughts!
Log in to join the discussion




Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime