Verdantix 2025 Global AI Survey: Three Takeaways For Corporate Leaders

In October 2025, Verdantix published the latest edition of its AI global corporate survey. Amidst market uncertainty regarding a potential bubble burst, the report details the budgets and priorities of 352 AI and IT leaders at large enterprises for 2026 and beyond. Three key narratives emerged from the findings:

• After years of aggressive double-digit expansion, enterprise AI budgets will enter a more measured growth period, centred on ROI.

  Our survey data indicate modest AI budget growth in 2026. More than half of respondents project a modest 1-9% annual AI budget increase and a further 26% expect no change at all. This represents a significant departure from 2025 projections, when only 28% of firms planned to make no change to budgets or increase spending by 1-9% – and a substantial majority predicted double-digit growth. The shift suggests that AI budgets have reached a mature baseline, enabling firms to pivot from scattered, low-impact pilots to scalable, high-value initiatives. As underperforming AI deployments are cut, organizations are reallocating capital to scaling the most successful projects with demonstrable ROI. Furthermore, as firms recognize that transformational AI requires deep structural changes to data and architecture, AI budgets are increasingly integrated into broader technical and functional budgets rather than defined in isolation.

• Large enterprises are bracing for widescale agentic automation by investing in infrastructural readiness and cyber security.

  Agentic AI is poised to hit the mainstream, with our survey indicating that most organizations will be trialling or piloting agents within at least one function in 2026. However, most corporate buyers expressed a need to focus on foundational readiness before scaling agents. Organizations are prioritizing the modernization of data infrastructure and technology stacks to enable adoption, often partnering with vendors like Databricks or Neo4j to build AI-focused data solutions such as ontologies and knowledge graphs. Cyber security and data privacy will be the most significant barrier to the AI rollout. As AI introduces new organizational risks – such as novel attack vectors and vulnerabilities in system embeddings – it heightens the need for robust security measures. Firms pursuing deeper AI integration are more likely to repatriate their workflows from the cloud back onto on-premise deployments. In response to these risks and others, Verdantix has developed a governance framework to benchmark readiness and assist firms in responsible AI deployments.

• AI is compelling firms to re-evaluate their enterprise tech stacks, forcing stick or twist decisions.
  As enterprises strategize how to embed AI across their ecosystems – weighing custom builds, verticalized solutions and AI-first agentic platforms – the priority is identifying vendors capable of facilitating the long-term, scalable transition toward deeper AI integration. Our survey indicates that most of the initial wave of AI integration will be deployed via updates to existing software. There is a preference to retain incumbent vendors within deeply embedded applications like ERP and CRM systems, as well as in low-ROI categories like BI. However, decision-makers are more open to nascent AI-first solutions for specialized software categories such as environment, health and safety (EHS), product lifecycle management (PLM) and supply chain management (SCM) software. Buyers want to pick the winning horse: a vendor that is resilient, has a well-articulated AI roadmap and business case, and can demonstrate progress in executing that vision.

Realizing ROI is a defining goal, with 2026 acting as the litmus test for the current iteration of LLM-enabled solutions for many organizations. To learn more about Verdantix market predictions and corporate priorities for the coming year, check out our AI Applied research or schedule an analyst inquiry.