Last week, Claude’s Fable 5 was disabled globally within hours, no migration window, no advance notice. Organisations that had built workflows around it had to figure out, in real time, how to keep operating.
It’s an extreme example. But the underlying question it raises isn’t extreme at all: if the tools your team depends on every day disappeared tomorrow, how long before your operations felt it?
For most organisations, the honest answer is: immediately.
The way many businesses have adopted AI over the last few years follows a familiar pattern. A team finds a tool that works. They build around it, automating a process here, accelerating a workflow there. It becomes load-bearing before anyone really decides it should be. And then it becomes invisible, right up until the moment it isn’t.
This is the quiet risk in AI adoption that doesn’t get talked about enough: whether the organisation could function without it. A few years ago, this wasn’t something anyone would have thought of, and now we wonder if teams are even asking this question.
Mature technology strategy has always been built on redundancy. Backup systems. Failover plans. The assumption that any single component can fail, and the discipline to plan for it before it does. AI deserves the same treatment, but because adoption has moved fast and often informally, that discipline hasn’t always followed.
The organisations that navigate disruption well, whether it’s a model going offline, a vendor changing terms, or a platform sunsetting a feature, share a few things in common:
- They know which of their workflows are AI-dependent.
- They have a clear picture of what breaks first and what has a fallback.
- And they’ve made deliberate choices about where AI is core to operations versus where it’s a nice-to-have.
In fact, the organisations moving fastest with AI are often the ones that have thought hardest about resilience, because they have more at stake. Speed and resilience aren’t in tension. They’re the same discipline applied at different moments. You move fast because you’ve done the work to know what you’re building on. You scale confidently because you’ve already asked what happens if a piece of it changes.
