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.
Sponsored By:
Insight AI
Archana Venkatraman
Senior Director Research, Cloud Data Management
IDC
Read bio
Archana Venkatraman
Archana Venkatraman is a Senior Director for IDC’s European Datacenter Research. She covers datacenter technologies including software-defined infrastructures, storage and data management, virtualization, containers, hyperconverged infrastructure, infrastructure performance monitoring, systems management, application development, and cloud services.
Archana also leads IDC’s European thought leadership program on open source technologies. She also contributes to European Digital Transformation, DevOps, Blockchain and IoT research practices.
Before joining IDC, Venkatraman was the datacenter editor at Computer Weekly, the digital magazine and website for IT professionals based in London, where she focused on datacenters, server virtualization, storage, open source technologies, software-defined infrastructures, and cloud computing, liaising with enterprise CIOs and technology vendors to develop deeper insight into the enterprise IT industry. Venkatraman has a master’s degree in journalism from Mumbai University.
Phil Hawkshaw
EMEA CTO
Insight
Read bio
Phil Hawkshaw
Phil Hawkshaw. EMEA CTO. An engineer, innovator and strategist. Helping organisations deliver technology transformation and adopt emerging technologies with confidence. Responsible for Insight’s strategic portfolio, AI innovation across EMEA and the UK Public Sector.
Event Sessions
One Day Event 11:45 am
The Amplification Effect: What Scaling AI Really Does to Cost, Risk, and Value
Across every industry, the same question keeps surfacing: how do we control AI cost and tie it back to the initiatives and outcomes it’s meant to serve? That question exists because scaling AI doesn’t just increase spend – it amplifies everything underneath, from duplicated work and governance gaps to an expanding attack surface that grows with every agent deployed. The conversation has shifted from maximising capability per use case to proving what each returns, with leading organisations already moving to smaller, purpose-built models that deliver precision at a fraction of the cost.
This session uses Insight’s AI Enterprise Architecture as the lens for making this visible, showing how AI FinOps, component reusability, and shared knowledge frameworks make it possible to scale AI with confidence, not just speed.