ASC 842 در آستانه وارد کردن ضربه‌ای سنگین به هیاهوی هوش مصنوعی است

Market impact: bearish for AI valuation optics, but unlikely to constitute a sudden fundamental shock.

The key distinction is that ASC 842 is not a new rule arriving in 2026 for U.S. public companies. Public entities generally adopted the standard for fiscal years beginning after December 15, 2018, and it already requires most operating leases to be recorded with a corresponding right-of-use asset and lease liability.

The more relevant issue is the future commencement of newly signed data-center, power, and infrastructure leases. Under ASC 842, a signed arrangement is generally not recorded as a lease liability until the lease commences. Consequently, rapid AI infrastructure expansion can create a pipeline of obligations that will appear progressively on corporate balance sheets rather than in one regulatory step-change. The article estimates substantial future obligations for MSFT, META, AMZN, and GOOGL, but those figures should be treated as author estimates rather than confirmed market-wide accounting adjustments.

For the hyperscalers, the immediate market concern is re-rating risk:

• Higher reported liabilities could worsen leverage, debt-adjusted valuation multiples, and return-on-invested-capital metrics.
• Investors may place greater emphasis on free cash flow, contractual commitments, capacity utilization, and AI revenue monetization rather than EBITDA or adjusted earnings.
• If lease commencements coincide with rising depreciation from accelerated GPU and data-center spending, reported earnings growth could face pressure even without a comparable near-term collapse in cash flow.
• The effect is most relevant to META and MSFT, where the article’s estimated lease exposure is large relative to the companies’ existing balance-sheet presentation; AMZN may attract less incremental shock because investors already analyze it with substantial infrastructure and lease obligations in mind.

However, the accounting change itself does not automatically destroy shareholder value or create equivalent funded debt. Operating-lease liabilities are accompanied by right-of-use assets, and the principal economic question is whether the leased capacity generates adequate returns. The market impact therefore depends on whether the disclosures reveal underutilized capacity, excessive fixed commitments, or weak AI monetization—not simply on the size of the liability line.

The most vulnerable trade is likely the broader AI-capex complex. If hyperscalers respond to balance-sheet scrutiny by slowing data-center expansion, delaying leases, or demanding better economics from suppliers, the pressure could extend to data-center operators, power providers, equipment vendors, and semiconductor suppliers such as NVDA. That transmission is a second-order risk, not an immediate consequence of ASC 842.

For data-center REITs and infrastructure lenders, the news is mixed. Greater lease capitalization may improve transparency around tenant obligations and credit quality, but it could also make hyperscalers less willing to sign very long contracts, reducing duration, visibility, and financing flexibility. This could raise the premium demanded for projects dependent on a small number of AI tenants.

Time horizon:

the initial effect is primarily sentiment- and multiple-driven. The medium-term impact will depend on lease commencement schedules, AI utilization, cloud pricing, depreciation policy, and whether revenue growth catches up with infrastructure spending. The bearish thesis weakens if hyperscalers demonstrate strong incremental AI returns, maintain robust free cash flow, and continue funding commitments without balance-sheet stress.

Traders should monitor:

upcoming 10-K and 10-Q lease disclosures, remaining performance obligations, purchase commitments, data-center construction timing, depreciation assumptions, free-cash-flow conversion, AI revenue growth, and any signs of capital-spending reductions. The central market question is not whether liabilities will be reported, but whether those liabilities represent productive capacity or excess fixed-cost exposure.