Recognize,a technology investment platform that focuses exclusively on the tech services industry, has shared some interesting findings from its recent CIO survey. The Recognize CIO Survey series is a periodic survey conducted with a panel of 200+ IT executives in the United States.
The data from this survey highlights how technology leaders are re-evaluating their IT strategies and budgets heading into 2026. The findings point to an environment of optimism and strategic recalibration, with most organizations doubling down on custom development, enterprise AI adoption, and managed services to drive innovation and efficiency.
The insights from the study can be grouped under four areas such as: (1) AI as a growth driver, (2) Enterprise AI maturity, (3) Governance and risk, and (4) The expanding services ecosystem.
Key Findings:
1) AI as a growth driver:
- IT budgets expected to expand heading into 2026: The survey reveals that a striking 85% of respondents expect their IT budgets to increase in 2026, reaffirming technology’s role as a business growth driver. Only 3% anticipate a reduction in budgets, while 9% expect spending to remain flat. This upward trend highlights growing confidence in digital investments, even amid economic uncertainty, as enterprises continue to modernize IT infrastructure and integrate AI-driven capabilities.
- Shift toward AI-generated enterprise applications over commercial software: A majority of respondents (55%) said they anticipate replacing some commercial software applications with AI-generated tools such as self-built CRMs and workflow automation platforms. Another 28% are considering this approach. This shift suggests that enterprises are beginning to view AI as a creator of intellectual property, not just a productivity enabler.About 10% either said no or were unsure, and 7% are currently evaluating the approach.
- IT teams expected to grow: In line with budget expansion and digital priorities, 67% of respondents’ organizations plan to increase the size of their IT teams over the next two years, while 28% expect to remain at current levels. Only 3% anticipate a reduction, underscoring the continued need for specialized tech talent.
- Accelerating adoption of AI and Large Language Models (LLMs): The data points to widespread experimentation and adoption of AI and Large Language Models (LLMs). 57% of respondents’ organizations already have major AI projects in production, while 55% are prototyping enterprise use cases and 48% report individual-level experimentation, suggesting both top-down and bottom-up momentum in AI adoption. Interestingly, only 3% reported not using AI at all, highlighting how deeply embedded AI has become across enterprise workflows.
2) Enterprise AI Maturity:
- Shift toward custom development for competitive differentiation: When asked about trends in proprietary application development, 42% of respondents said they are accelerating custom development to strengthen competitive differentiation. In contrast, 36% are reducing custom builds in favour of commercial off-the-shelf or SaaS solutions for faster deployment and cost efficiency. The remaining 22% reported no major change in their current approach. These figures indicate a two-speed IT model emerging across industries, with some doubling down on innovation, while others prioritize stability and standardization.
- Fine tuning and LLM customisation on the rise: Half of all organizations surveyed (50%) have already begun fine-tuning commercial LLMs to suit their enterprise needs, while another 37% are still evaluating such initiatives. This demonstrates growing interest in building contextually relevant AI systems instead of relying solely on off-the-shelf models.
3) Governance and Risk:
- Security tops IT leader concerns in accelerating Gen AI deployment: The survey sheds light on major concerns that organizations have regarding the rapid deployment of Generative AI. The biggest concern is around security, cited by 64% of respondents, highlighting the pressing need for robust AI governance frameworks.
- Other concerns include performance and latency (37%), complexity (35%), costs (34%), and inaccurate results (33%). Sourcing skilled talent (27%) and job impact (24%) were also cited as challenges, indicating that while AI excitement is high, readiness and risk mitigation remain key focus areas. Additionally, 9% worry about low ROI, and 3% are unsure about relevant use cases.
4) Expanding services ecosystem:
- Contractor ecosystem expands with AI integration: With AI reshaping enterprise operations, 54% of respondents said their organizations plan to increase their use of system integrators and contractors to build and manage AI infrastructure and applications. Another 29% expect to maintain current levels but shift focus toward AI, signalling a healthy demand environment for external expertise.
- Managed services play a pivotal role in AI deployment: As AI deployment matures, 64% of respondents said their organizations are partnering with third parties to manage the inference layer, handling performance, monitoring, and cost optimization, while 17% manage it in-house and 19% are still evaluating. This points to a growing reliance on specialized managed services in AI infrastructure management.
- OpenAI and Google lead the LLM race: When asked about preferred LLMs, OpenAI (90%) and Google (82%) emerged as the top two platforms of choice, followed by Meta (53%), Anthropic (27%), and Mistral (10%). The dominance of these players underscores how established ecosystems continue to influence enterprise AI decisions. The responses also underline OpenAI’s leadership in the enterprise AI space, even as competition among LLM providers continues to intensify.
- Wide range of AI coding assistants being adopted: Development teams are actively experimenting with a range of AI-powered coding tools. The most widely used are GitHub Copilot (55%), OpenAI Codex (54%), and Gemini CLI (49%), followed by Amazon Q Developer (40%), CodeGPT (43%), and Claude Code (34%). This widespread usage reflects how AI has become embedded in software engineering, helping developers boost productivity and accelerate release cycles.