.png)
Chart 1
Source: IDC
Chart 3
Source: S&P
On the utilities side, financial leverage could drift up modestly, as AI-driven data-center load will lift electricity grid and generation capex. Credit stability will hinge on timely regulatory cost pass-through mechanism. Markets with liquid power hubs and faster interconnection timelines will capture a larger share of growth. Where power can be priced, procured, and interconnected quickly, data-center builders will concentrate, and so will growth in utility capex, load, and credit-supportive investment.
In data-intensive, process-driven sectors, finance, healthcare, insurance, and logistics, AI agents can execute complex tasks in seconds by automating standardized workflows and extracting signals from large datasets.
In finance, for example, production deployments of AI-driven fraud screening have, in some cases, cut false positives by over 50%, reducing manual review, compliance costs, and loss ratios.
In healthcare, agents streamline prior authorizations, clinical coding, and discharge planning; in insurance, they accelerate claims triage and sharpen pricing.
In logistics, they enable dynamic routing and demand-aware inventory forecasting. From a credit perspective, these use cases lift throughput and margins with relatively modest incremental capex.
Table 2
Source: S&P
Media, advertising, and slower-moving software firms will face the sharpest near-term risk. Increasing disruption of AI-powered chatbots for information retrieval are changing users’ behaviours.
AI contributed to an estimated 15% decline in global search traffic year to date as of June 2025, according to Similarweb data. eMarketer reported that Google AI overviews led to as much as 52% decrease in referral traffic in a single month and are reducing users click-through behaviour.
If online traffic declines for publishers persist, advertising monetization reduces meaningfully, these could eventually pressure publishers’ credit profile and repayment capacity.
The software sector is undergoing a structural reset. AI is lowering barriers to entry and accelerating product cycles, shifting demand across categories. Pricing power and market share are eroding for undifferentiated vendors, and margins are compressing for laggards.
For lower-rated issuers, intensifying competition and higher customer-acquisition costs, just as capex rises elsewhere in the stack, will squeeze free cash flow and rating headroom.
Conversely, large and highly rated software companies with scalable platforms that have credibly embed AI should realise structural efficiency gains. These companies can unlocked new revenue models – such as AI-ed product enhancements and usage-based pricing – that will strengthen customer retention and cash generation (Table 4).
Table 4 : Software Companies are rapidly adopting AI
Source: Fitch
Regional divergence seems inevitable.
The United States is already pulling ahead, with recent productivity gains, estimated by S&P to have lifted potential growth by 0.4% to 0.5%, driven by sustained investment in energy transition technologies and the rapid diffusion of AI. The CHIPS Act has amplified this trend, funnelling subsidies into domestic semiconductor capacity, from fabs and tools to advanced packaging, laying the hardware foundation for scalable AI adoption.
Credit resilience will naturally cluster in economies with abundant power supply, faster permitting processes, and deep capital markets, conditions that favour the U.S. above emerging markets.
By contrast, many emerging market companies face tighter financing conditions and higher capital costs, stretching the timeline for AI payoffs.
Emerging-market liquidity has greatly improved in recent months, buoyed by a softer U.S. dollar and investors’ search for yield. Yet borrowing costs remain elevated. Should global growth lose momentum, and the dollar strengthen, the tide could swiftly turn. External shocks could test underlying vulnerabilities, tightening financing access and leaving emerging market borrowers once again facing funding strain.
Compounding the risk, most emerging markets are net energy importers (Chart 5). Higher energy prices could prompt emerging market central banks to recalibrate policy, slowing or pausing monetary easing cycles, and tighten funding conditions for corporates.
Chart 5
Source: S&P
A sharp pullback in AI infrastructure spending could hit technology hardware revenues hard and erode rating headroom – falling profit margins, weaker cash flow, and tighter FFO-to-debt (FFO: fund from operating cash flow).
In a stress scenario modelled by Fitch, assuming a 50% collapse in AI-related revenues by 2027, experience a temporary spike in agency’s adjusted leverage (debt-to-EBITDA), exceeding their leverage downgrade trigger (red in Table 6). Such breaches rarely prompt instant downgrades. While leverage is a key trigger, rating agencies weigh a wider mix of financial metrics and qualitative factors before acting. More importantly, ratings reflect corporate credit quality over the business cycle, not the noise of short-term earnings swings.
Table 6
Source: Fitch, 2025E leverage is based on Fitch forecast for each’s next fiscal year
From 2026 onward, agencies expect AI to be a measurable differentiator in credit quality. Winners will show productivity gains, pricing power and operating expense efficiency that lift margins and stabilize cash flow, improving debt-service capacity and opening the door to upgrades. Laggards (over-spending, mis-executing, or falling behind) risk free cash flow erosion, higher leverage and potential downgrades. By 2027, agencies anticipate AI readiness to be embedded into sector scorecards under “business risk” and “financial policy” factors, moving it from narrative to numeric driver.