How Fortune 500 Leaders Are Really Responding to AI Disruption (What Boardroom Conversations Look Like in 2026)
There's the version of this conversation that happens at conferences, in press releases, and in the public statements of CEOs who want to be seen leading on AI. And then there's the version that happens in boardrooms and executive briefings when the cameras are off.
They're different conversations.
After 12 years at Deloitte Consulting as Senior Manager, Strategy & Innovation working directly with leadership teams at Fortune 500 organisations including Walmart, PwC, Morgan Stanley, Johnson & Johnson, and Pfizer and continuing that advisory work through Queen & Rook and Moonstone AI, I've been in both versions. This is as honest an account as I can give of where executive leadership actually is on AI in 2026.
The gap between public positioning and private uncertainty
Most large organisations have issued some form of public statement about their AI strategy. They've appointed a Chief AI Officer, stood up a centre of excellence, and announced a multi-year investment. In their earnings calls and annual reports, AI is presented as a strategic priority with clear direction.
In the actual boardroom, the conversation is more honest. The questions I hear most frequently from senior leaders are not about which AI tools to adopt. They're about accountability, governance, and risk.
The accountability question
When an AI system makes a decision that turns out to be wrong, a credit approval, a hiring recommendation, a supply chain call, who is responsible? This is not a hypothetical question. Organisations deploying Agentic AI at scale are already encountering this in operational contexts, and most have not resolved the governance question satisfactorily.
The leaders navigating this well are those who have separated the question of automation from the question of accountability. They're building human review into the loops that matter most not as a compliance exercise, but as genuine risk management.
The trust gap in the workforce
One of the most consistent themes from CHROs and COOs is that the workforce is not where leadership assumed it would be on AI adoption. The assumption was that employees would embrace AI tools that made their jobs easier. The reality is more complicated.
People are not resisting AI because they don't understand it. They're resisting it because they can't see where they fit in the picture it's painting.
The board's AI literacy problem
Most corporate boards are not equipped to provide meaningful oversight of AI strategy. Board-level AI conversations tend to default to one of two modes: either uncritical endorsement of whatever management presents, or reflexive risk aversion triggered by whatever AI story was in the news that week. The organisations handling this best are investing in board education sector-specific briefings on the specific decisions the board will need to evaluate over the next 18 months.
What the leading organisations are actually doing differently
They've separated AI experimentation from AI strategy
Experimentation, running pilots, testing tools, exploring use cases is valuable and necessary. But it is not a strategy. The organisations that are ahead have made a clear distinction between 'we are learning what AI can do' and 'we have decided what role AI will play in our competitive model.'
They're treating Agentic AI as an organisational design question
The shift from generative AI to Agentic AI is not primarily a technology question, it is an organisational design question. When software can autonomously execute multi-step workflows, the structure of teams, the nature of managerial roles, and the flow of information through an organisation all change. The Innovation in a World of AI keynote goes deep on exactly this question for leadership audiences.
They've invested in AI governance before they needed it
Key governance questions that well-prepared organisations have already answered:
Which decisions can AI make autonomously, which require human review, and which require human decision with AI support?
How are AI-generated outputs audited and by whom?
What is the escalation path when an AI system produces an output that conflicts with company policy or legal requirements?
How is training data governed, and what are the privacy implications of using internal data to fine-tune models?
The industries where executive posture is most different from public positioning
Financial services
Financial services organisations have more AI deployed in production than almost any other sector. The executive conversation in 2026 is not about whether to use AI, but about how to govern Agentic AI systems that are increasingly making consequential decisions at speeds that preclude meaningful human review.
Healthcare
Healthcare executives are navigating a genuine tension between the proven value of AI in clinical decision support and the regulatory environment that has not yet caught up to the pace of AI capability. The boardroom conversation is often about liability specifically, the gap between what AI can do and what can be defended in a regulatory or litigation context.
Government
Public sector leadership is often further along in AI thinking than it gets credit for. The challenge is procurement and implementation speed, not strategic understanding. The Fearless Government keynote addresses exactly this how public sector organisations move faster inside constraints.
What executives are getting wrong
Treating AI adoption as a technology initiative rather than a change management challenge. Technology is the easier part.
Assuming that a Chief AI Officer appointment resolves the governance question. It doesn't, it just moves the question to one person's desk.
Measuring AI success with the wrong metrics. Efficiency gains are measurable but capture only part of the value and none of the risk.
Underestimating the speed of the Agentic AI transition. Most organisations are planning on a two to three year runway. The timeline is shorter than that.
For organisations wanting to bring this conversation to their leadership team, the Innovation in a World of AI keynote is specifically designed for executive and board-level audiences. You can also read more about Shawn's background working with Fortune 500 leadership on the Meet Shawn page.
Frequently asked questions
How are Fortune 500 boards actually overseeing AI strategy in 2026?
Most boards are still developing the oversight frameworks they need. The leading practice is establishing a dedicated AI risk committee or expanding the remit of the technology or audit committee to include AI-specific oversight. The key gap is board-level AI literacy and the organisations addressing this are investing in sector-specific education for board members, not generic AI briefings.
What is the most common mistake large organisations are making with AI adoption?
Treating AI adoption as a technology initiative. The technology is solvable. The harder problems are change management, governance, and the workforce trust gap. Organisations that frame AI adoption as a technology project consistently underinvest in the cultural and structural changes that determine whether the technology actually produces value.
How are organisations handling the cybersecurity risks of Agentic AI?
The specific risk profile of Agentic AI systems that can take autonomous actions, access external services, and operate across connected workflows is different from earlier AI applications. The leading organisations are applying the principle of least privilege to AI agents, maintaining detailed audit logs of agent actions, and building explicit human review into any agent workflow that touches sensitive data or consequential decisions.
What does a mature AI governance structure look like inside a large organisation?
At minimum: a clear policy on which decisions AI can make autonomously versus which require human review, a defined process for auditing AI outputs, an escalation path for AI outputs that conflict with policy or legal requirements, and a data governance framework that covers how internal data is used in AI training and inference.