Today’s ponderable: When did AI governance stop being a constraint and start becoming a condition of progress?

Dear Diary

As a JPMorgan Chase alumna, I continue to follow the company and its executives with interest. This is partly personal, of course – but it is also because JPMorgan Chase remains one of the clearest examples of how a large, complex enterprise is turning AI ambition into governed, operational reality.

In recent months, I have paid especially close attention when JPMorgan Chase – and in particular its CEO, Jamie Dimon – talk about AI. The company is not debating AI from the sidelines; it is living the realities of scale, risk and accountability in real time. And that makes it a useful lens for a broader pattern now playing out in boardrooms: AI is landing as innovation because it is fast-moving, high-impact and full of promise. For a while, it sits comfortably in this category. Something to explore, pilot and scale where it works.

The uncomfortable truth: scale forces the issue

Included in the DWG 2026 predictions for the digital workplace (Prediction 10) was my belief that very quickly AI executives would begin to move on from what AI can do to focus instead on a different set of questions: What happens when AI goes wrong? Who is accountable? How far can the organization push before risk catches up?

This very quickly proved to be the case and governance shifted from an abstract principle to a practical necessity. At JPMorgan, this shift did not happen because AI was novel. It happened because AI reached scale.

Embedded across fraud detection, risk modelling, customer operations and internal workflows, AI moved beyond experimentation into something that could materially shape outcomes: financial outcomes; customer outcomes; regulatory exposure. At that point, governance changes character. It went from a level of procedural hygiene to a matter for strategic oversight.

This is consistent with Jamie Dimon’s public stance. AI is not a side story; it is foundational, touching virtually every function across the enterprise.¹

Once you accept that premise, the next step is unavoidable. You cannot scale AI without scaling accountability alongside it.

The leadership tension: discipline over reaction

Dimon has been unusually clear about the risks of getting this wrong.

Too much control, applied too early, and organizations slow themselves down. Innovation fragments and momentum dissipates. In his words, there is a real danger in regulating so aggressively that you erase the value you are trying to capture.¹

Too little control and the inverse happens. Risks accumulate quietly until they surface in ways that are harder to contain. Trust erodes and regulatory scrutiny intensifies. Both paths are visible in today’s market.

The more interesting question is what sits between them. It is not balance in the abstract but discipline in execution, requiring preparation before incidents occur and honest assessment when they do. This demands a willingness to fix what breaks without dismantling what works.¹

That posture shows up less in policy documents and more in how decisions get made under pressure.

What this looks like in practice

Across JPMorgan and other large enterprises, a pattern is emerging:

1. AI is treated as infrastructure, not experimentation. It sits inside core processes rather than at the edge.
2. Governance builds on existing systems. Risk committees, audit structures and model validation frameworks extend into AI rather than being replaced by it. In financial services in particular, AI oversight is folded into established model risk disciplines, bringing it under familiar forms of scrutiny.²
3. Governance evolves. Static controls do not hold in environments where the technology itself is learning and adapting. What does hold is an operating model that expects iteration. Test early. Learn quickly. Adjust.
4. The pace of change is managed with intent. Not just within the organization, but in relation to the workforce and wider society. Dimon has been consistent here: AI will deliver significant benefits, but only if it is introduced in a way that does not leave people behind or destabilize the system around it.³

The shift that matters: governance expands the frontier

There is still a tendency to treat governance as a drag on progress. In practice, it is often the opposite.

Weak governance limits how far organizations are willing to go. It introduces hesitation and creates exposure that leadership teams are reluctant to carry forward. Strong governance does something more interesting. It increases the range within which organizations can operate confidently, allowing them to deploy AI more deeply, across more functions, with fewer reversals. This is where the competitive dynamic starts to shift.

AI capability in its own right is not scarce. Every major enterprise has access to similar tools. The differentiator is operational confidence – that’s the ability to move quickly and sustainably at the same time.

Closing thought

The conversation about AI governance is still catching up with reality. With enterprise AI reaching scale, the question ‘du jour’ is whether governance is keeping pace with ambition.

In this next phase, organizations that lead will not be defined by how fast they adopt AI or how tightly they control it. They will be defined by whether they can do both, without one undermining the other.

References

1. Jamie Dimon says AI will impact ‘virtually every function’ at JPMorgan Chase (Yahoo! Finance, Apr 6, 2026)
2. JPMorgan Chase Board Committees (JPMorgan Chase, 2026)
3. Jamie Dimon’s nuanced view on AI’s economic impact (IndexBox, Dec 7, 2025)

Categorised in:   → Diary of a She-E-O