Artificial intelligence (AI) is set to fundamentally alter corporate business models, according to a report from Fitch Ratings released on Wednesday. However, the rating agency cautioned that credit risks associated with AI adoption are heavily concentrated in specific sectors, notably technology, media, and telecommunications (TMT).
Fitch identified disruption and overinvestment risks as the main pathways through which AI could affect corporate credit metrics. Despite these risks, the agency does not anticipate immediate changes to ratings across most sectors. Overinvestment risk refers to the potential for companies to allocate excessive capital to AI infrastructure without achieving proportional returns, which could potentially weaken their credit profiles.
Disruption risk, on the other hand, pertains to the threats posed to established business models as AI facilitates the emergence of new competitors or substitutes. Both types of risk are particularly acute in asset-light industries, such as software, media, and services, where value is often derived from intangible assets including intellectual property and proprietary data.
The report highlighted that the AI-driven capital expenditure (capex) surge is largely concentrated among a few hyperscalers and cloud service providers. Fitch noted that the “big four” — Alphabet, Microsoft, Amazon, and Meta — are expected to invest approximately $650 billion in AI-related infrastructure by 2026, nearly matching their cumulative spending from 2020 to 2024.
Despite these significant figures, Fitch observed that the broader spectrum of rated corporates is adopting a cautious approach, with capital expenditures primarily focused on addressing visible demand rather than aggressive growth strategies. Data from Fitch’s Global Corporate Cash Flow Monitor, which includes over 1,500 non-financial issuers, indicates that capex intensity for North American corporations (excluding hyperscalers) is projected to increase slightly to 7.4 percent of revenue in 2025–26, compared to 6 percent to 7 percent over the previous five years. This uptick is largely backed by robust operating cash flows and is not expected to significantly impact free cash flow.
Outside the TMT sector, power utilities rank among the largest capital spenders, driven by essential needs such as aging infrastructure, renewable energy integration, and grid resilience rather than direct AI investments. Fitch reported that utilities typically refrain from committing to long-term capacity expansions solely to accommodate AI-driven demand, focusing instead on prudent investments that maintain balance-sheet flexibility.
Moreover, Fitch pointed out that ongoing supply chain disruptions affecting semiconductors, data centers, and power grids could hinder AI capacity expansion. Current lead times for memory, storage, and semiconductor fabrication remain constrained, while delays in power grid interconnections and permitting processes are extending timelines for new data-center connections. These challenges may result in elevated capex among hyperscalers, even if the pace of capacity delivery is slower, as companies might incur higher costs for expedited procurement.
While overinvestment risks are particularly prominent among hyperscalers, Fitch believes that disruption risks are predominantly found in asset-light sectors where AI can readily substitute human labor or existing processes. The agency pinpointed software, services, and media as the most vulnerable industries.
In the software sector, Fitch noted that AI is likely to accelerate the development of straightforward applications, such as workflow automation and customer interfaces. However, it emphasized that enterprise software featuring mission-critical functions and high switching costs remains relatively insulated, especially in regulated industries like healthcare and finance.
In terms of services, the agency highlighted that outsourcing for customer service, data analytics, and other repetitive tasks may face significant displacement risks due to AI advancements. Conversely, firms with proprietary data, entrenched ecosystems, long-term contracts, or regulatory protections are better positioned to withstand rapid industry changes.
Regarding the media sector, Fitch observed that AI-generated content has the potential to lower production costs. Although fully AI-generated content has yet to achieve studio-quality standards, tools such as automated summaries have begun to affect digital advertising revenues and search traffic for publishers.
Fitch also noted that resilience tends to be higher among companies with mission-critical offerings, regulatory integration, proprietary data, and financial flexibility, allowing them to invest in AI capabilities and manage transitional costs effectively. Despite the pervasive interest in AI, Fitch cautioned that transformative positive effects on corporate ratings are likely to remain limited for the foreseeable future.
While AI could enhance efficiency and reduce costs across various sectors, Fitch stated that revenue opportunities tied to AI adoption are still in their infancy and predominantly incremental. For example, pharmaceutical and media companies may see faster research and development cycles, while retailers could leverage AI for more targeted marketing. However, these incremental benefits are challenging to quantify and unlikely to sway ratings in the near term.
Even within sectors that are particularly exposed to AI, such as semiconductors and cloud services, Fitch emphasized that diversification in end markets and customer bases helps to mitigate the downside credit risks associated with a potential slowdown in AI advancements.
See also
Meta Unveils Four New MTIA Chips with 25x Compute Gains for AI Inference
Tesseract Launches Site Manager and PRISM Vision Badge for Job Site Clarity
Affordable Android Smartwatches That Offer Great Value and Features
Russia”s AIDOL Robot Stumbles During Debut in Moscow
AI Technology Revolutionizes Meat Processing at Cargill Slaughterhouse
