Track 4

Value is Bespoke: Guardrails, Not Gatekeepers – Governing AI Investment & Growth By Design

2026 is an inflection point with enterprises shifting from initiating their artificial intelligence (AI) pivot and start to make investments in more advanced multi-agentic AI systems. 50% of all organisations globally and 74% of “digital native” organisations are using agents in production across multiple business areas, IDC research shows.

As AI moves rapidly from experimentation to enterprise execution and at machine scale, a new challenge is emerging: how do you govern that growth in a way that protects investment without becoming the ceiling on innovation?

IDC predicts that by 2027, G2000 agent use will jump 10x and token/call loads 1,000x, making agent vetting, orchestration, and optimisation essential IT responsibilities. Already, in conversations with IDC, organisations admit exhausting their full year’s AI budgets in the first six months with limited demonstrable ROI.

The answer is more nuanced than just a one-size-fits-all framework. A generalised framework that works for one organisation may not necessarily yield the same results for another organisation as it depends on multiple factors including investments, governance maturity, stakeholder backing, data maturity, use-case alignment and most importantly organisational context and value metrics.

Value in an AI-driven enterprise is not universal. Neither are all the metrics from traditional digital enterprise fit for purpose to measure AI-driven business value. A financial services organisation managing regulatory exposure, a manufacturer optimising physical operations, and a professional services firm scaling knowledge delivery face fundamentally different challenges.

IDC research shows only 18% of organisations said that most of their AI-related projects delivered measurable business outcomes. When it came to all AI initiatives delivering business outcomes, it was just 3% of respondents.