AI revenue at Meta, Alphabet, and Microsoft tops $25 billion, edging past AI infrastructure depreciation costs for the first time.
Meta's AI revenue, driven by enhanced ad targeting and content recommendation, has grown enough to cover depreciation on its AI infrastructure. This suggests its heavy investment in AI chips and data centers is beginning to yield tangible financial returns, though the slim margin leaves little room for error.
It reduces the net cost of AI infrastructure on earnings, as revenue offsets a portion of the non-cash depreciation charge. This can improve reported operating margins, but the narrow overage means profits remain highly sensitive to any revenue deceleration.
Further scaling of AI-powered advertising in emerging markets and deeper integration of AI into messaging platforms like WhatsApp could accelerate revenue, while any new AI-driven product launches may expand the monetizable base.
Alphabet's AI revenue from Google Cloud and its Gemini-powered advertising tools reached a level that exceeds depreciation costs of its AI servers. The milestone indicates that its AI push is starting to pay for itself, but the narrow gap highlights ongoing cost pressures.
It shows that Google Cloud's AI services are generating enough revenue to cover the depreciation on related hardware, a sign that the cloud division's AI pivot is gaining commercial traction beyond traditional workloads.
Increased competition from AWS and Azure in AI services, or a slowdown in enterprise AI adoption, could keep revenue growth below the level needed to comfortably outpace rising depreciation from continued infrastructure expansion.
Microsoft's AI revenue, led by Azure AI services and Copilot integration, surpassed depreciation expenses for the first time. While this validates the early monetization strategy, the minimal margin underscores the heavy upfront capital burden.
It suggests that the heavy investment in OpenAI and AI infrastructure is beginning to generate returns that at least cover the depreciating hardware costs, a positive signal for the long-term viability of Copilot and Azure AI.
Yes, as more enterprises adopt AI copilots and Azure AI services, revenue could compound. However, the pace of depreciation will also rise with continued data center buildout, so the spread between revenue and depreciation will be a key metric to watch.
AI-derived revenue for Meta, Alphabet, and Microsoft reached $25 billion in the latest quarter, just exceeding depreciation on AI-related capital spending. The narrow margin highlights both the rapid scaling of AI businesses and the immense infrastructure costs yet to be recouped. While crossing this threshold signals that AI investments are starting to pay off, it underscores a fragile balance that could shift with any deceleration in growth.
It indicates that the revenue generated from AI products and services is now greater than the accounting cost of wear and tear on the hardware and data centers required to support AI. This is a key milestone in validating the return on investment for massive AI infrastructure spending.
Investors have been watching whether Big Tech's multi-billion dollar AI capex will ever generate proportional returns. Surpassing depreciation suggests the spending could become self-funding, reducing the risk of margin compression.
No, exceeding depreciation is just one step. Depreciation is a non-cash expense that allocates the cost of physical assets over time. Full profitability would require covering all operating expenses and generating a surplus, which is still a distant goal.