We're already a few weeks into 2026, and the pace of AI developments continues to accelerate. The focus of discussions around AI is shifting decisively within organisations – from what is technically possible, to what is operationally viable at scale. Leaders are moving beyond isolated pilots to confront more complex challenges around governance, trust, and end-to-end integration.
As 2026 gets underway, we asked some of McKinsey’s technical leaders to share the themes they believe will shape the year ahead, and where they see the greatest opportunities for UK organisations:
Debasish Patnaik, Senior Partner, leads QuantumBlack, AI by McKinsey for the UK
“Everyone expects 2026 to be the year of fully autonomous agents; when what is more likely is it’s the year autonomy shrinks by design. Enterprises will move from “AI that chats” to “AI that settles”, closing tickets and clearing exceptions end-to-end with audit trails, and the real productivity lift will come from automating coordination (handoffs, approvals, escalation), not just tasks. The business value of AI will correlate less with how much you spend on AI models, and more on how well organisations redesign everyday workflows that see AI used in controlled environments with verifiable actions and hard guardrails.”
Kate Smaje, Senior Partner and Global Leader, Technology and AI
“We’re seeing a real step up in engagement and ownership around AI by boards and C-suite teams. They're starting to own end-to-end outcomes rather than delegating experiments and individual use cases. 2026 should be the year organisations convert 2025’s pilots into production at scale, with evaluation, controls, and the “humanware” on top that makes the impact material and repeatable.”
Alex Sukharevsky, Senior Partner, AI and digital leader
“In 2026, many enterprises will realise that while AI technology is advancing fast, significant value won’t materialise without the fundamentals – modern IT architecture, high-quality data, capabilities, operating model, and change management. We’ll see a tougher “audit moment” where programmes fall short not because the models underperform, but because the enablers and economics weren’t in place. The gap between AI natives and traditional organisations will become more visible, competitive dynamics will shift, and the focus will move from scaling AI to secure, ethical scaling built on digital trust.”
David Champagne, Senior Partner, leads QuantumBlack in Life Sciences
“2026 will bring a few wake-up calls for corporates: a company in an industry will successfully scale agentic workflows and create a step-change in speed, cost, or service levels. When that advantage is visible, it will jolt everyone else from experimentation to production.”
Dave Kerr, Partner, leads Agentic deployments
“The focus will shift from “can it work?” to “can we trust it, consistently, in the real world?” This year I expect to see real confidence built in AI through trust and assurance. More autonomous AI systems will start being used widely in real business operations once organisations can prove they are reliable, behave consistently, and have strong controls in place to prevent problems. LLM-as-judge is a useful tool, but it should sit within a broader approach that combines deterministic tests, statistical measures of accuracy, and robust safeguards.”
Rory Walsh, Partner, leader within QuantumBlack Labs
“This will be the year of enterprises tackling the foundational enablers required to build, govern, and operate Agentic workflows in a way that meets business needs, security obligations and regulatory requirements. Moving beyond proof of concepts and single deployments will require considerable investment. Key topics will include at-scale evaluations, agentic observability across 1st and 3rd party agentic platforms and Identity and Access Management for agents.”
Emily Sagar-Jones, Principal Data Scientist at QuantumBlack
“Looking ahead to 2026, I’m most excited about the idea of the Agentic OS becoming a reality – an operating layer that anticipates needs, responds to events, and coordinates across systems by deploying the right agents autonomously in the background of everyday work, escalating to humans only when necessary. This shift will increase the level of autonomy entrusted to AI, while placing importance on robust security, governance, and guardrails.”
Collectively, these perspectives point to a clear differentiator in 2026: execution. Competitive advantage will depend on a sustained investment in fundamentals rather than incremental experimentation. Organisations that hard-wire AI into redesigned processes, build confidence through robust governance and controls, and invest in the human and technical foundations will be better positioned to translate AI capabilities into performance gains.
This article was compiled by Rohit Rathod, a Senior Client Development Advisor in our London office. For McKinsey’s latest thinking on how organisations can most effectively and responsibly use AI to create business value, explore our Insights on Artificial Intelligence.