A recent survey by OutSystems, The State of AI Development 2026, reveals that AI has entered the early production phase for many enterprises, primarily within the IT function. The survey, which gathered responses from 1,879 IT leaders, warns that AI adoption is at risk of outpacing governance and integration.
The report highlights a significant gap between what IT leaders want AI agents to achieve and what their organizations can safely control. To address this shortfall, the report’s authors emphasize the need for companies to establish controls or guardrails on AI systems and integrate new AI technology into existing platforms.
Notably, 97% of respondents are exploring agentic strategies, with 49% describing their current abilities as ‘advanced’ or ‘expert.’ Nearly half of those surveyed reported that over half of their agentic AI projects have moved from pilot to production, with Indian companies leading the way in implementing the technology.
While ‘cost reduction or efficiency gains’ is the most cited expectation for AI’s effects, only 22% of respondents found their deployments most effective in this regard. Instead, the most effective area of gains in a business stemmed from equipping software developers with AI tools, described as ‘generative AI-assisted.’
The report’s findings also show that transitions to AI agentic workflows are unevenly distributed across geographies and sectors. India stands out as the market with the highest share of users considering themselves ‘expert,’ while many organizations in other countries still identify as intermediate stage users.
The financial services and technology sectors show the most movement from pilot to production, with many implementations in core business functions. The report suggests that slower-moving sectors can learn from the implementation workflows employed by the fintech industry, starting with narrow, high-volume workflows where performance can be measured and failures can be contained.
Photo by Mikhail Nilov on Pexels
Photos provided by Pexels