EY released its top ten opportunities for tech companies in 2026 on December 15, 2025

On December 15, 2025, EY published its annual report The Top 10 Opportunities for Technology Companies in 2026. The publication offers recommendations for firms in the sector and stems from three pressures: AI adoption, geopolitical complexity, and the need to demonstrate economic results. It does not announce a new product or a concrete investment by EY in the industry.

The document is signed by James Brundage, Ken Englund and Andrew Young, who are linked to EY’s technology sector analysis. The page indicates a 15‑minute read time and presents the text as a guide for actions that drive growth, resilience and confidence. Its date matters: it is a perspective published at the end of 2025 to inform decisions in 2026, not a news item today or a recap of results already observed over the year.

The idea that organizes the report is that technology firms must demonstrate ROI on their AI investments and take projects beyond isolated pilots. EY argues that the challenge is no longer just incorporating models or agents, but integrating them into operations and commercial offerings with controls, interoperability and a measurable outcome. The practical shift is in scale and accountability, not merely technological adoption.

EY also describes an environment it calls NAVI—non‑linear, accelerated, volatile, interconnected—and maintains that this combination increases pressure on strategy and execution. It is EY’s interpretive framework, not an independent market metric. For a CIO, the useful reading is to ask whether each initiative has a responsible person, a baseline, an expected outcome and security criteria before expanding its use.

Among the first opportunities, EY proposes using acquisitions and joint ventures to quickly access capabilities, markets and data assets. The document cites an EY survey of technology CEOs in which 83 % would prioritize partnerships and joint ventures in the coming months. That percentage reflects a stated intention, not closed deals or a growth forecast; it also comes from a survey conducted by the consulting firm itself and should be read with that scope.

Another recommendation is to design AI products so they work across platforms and clouds, rather than relying on a single ecosystem. EY links this interoperability to agentic AI and physical AI, where software systems can coordinate actions and, in some cases, interact with machines or industrial environments. For product teams, the consequence is to evaluate interfaces, data portability, permissions and human oversight from design onward; the report does not set a concrete technical standard that solves those decisions on its own.

The report recommends training functional leaders to broaden secure and reliable AI, with governance responsibilities embedded in deployment. The proposal acknowledges an operational tension: accelerate production without letting security and accountability become late reviews. Governance becomes part of the product and process, not a task exclusive to legal or security teams.

EY adds default sovereignty as a principle for deciding where data, models and workloads are hosted and processed. In practice, this means reviewing jurisdictions, contracts, vendor dependencies and access requirements before adopting multi‑cloud services. It does not mean all firms must abandon public cloud: the report proposes a strategic guideline, while the appropriate architecture depends on regulation, data and risk in each organization.

Among remaining opportunities, EY suggests deploying technical specialists so AI capabilities reach more teams and strengthen governance as uses mature. It also proposes reviewing commercial models, including performance‑linked pricing, and optimizing AI across functions. The decisive link is between technology and the business economy: a technical improvement only becomes an advantage if the customer values it and the firm can sustain its cost.

This requires finance and product to agree on how to measure benefits, inference costs, integration, support and risk. Instead of counting pilots or users as sufficient proof of success, a company can define indicators tied to savings, revenue, quality or operating time, depending on the case. The report does not prescribe a universal formula nor guarantees that charging for results is suitable in all markets; the recommendation requires contracts that allow attributing outcomes with clarity.

For executives and CIOs, the value of the report lies in using it as a list of questions to prioritize investments, not as an obligatory roadmap. A concrete application is to compare existing initiatives against four conditions: measurable value, interoperability, security and operational capability. If an initiative lacks metrics or responsibility, expanding its deployment may raise costs and risks before proving usefulness.

An executive review can start like this:

• Identify which AI projects are in production and what business outcome they measure.
• Check vendor dependencies, portability and data rights.
• Assign security, oversight and incident‑response owners.
• Evaluate whether a partnership or acquisition adds a capability that is not worth developing internally.
• Review if the revenue model matches the value received by the customer.

These checks do not replace financial, technical or regulatory analysis. They help separate a strategic priority from a trend and decide what to experiment with, scale or stop.

The 2026 report should be read alongside its previous editions, not as a set of comparable results alone. The 2025 edition, published on December 18, 2024, focused on turning AI promise into value and transforming operations; the 2026 edition emphasizes partnerships, interoperability, governance and outcomes. The shift suggests an editorial evolution at EY, but does not itself prove that the market has adopted those practices or generated returns.