When I lost access to Fable, I had a backup plan. Now that it's back, I'm even more certain that my fail

This as-told-to essay is based on a conversation with Sean McDonnell, 43, who lives in England. McDonnell is the founder of the web design company Kaizen and the SaaS website Consigns. The following has been edited for length and clarity.

Last month, a few hours into trying Anthropic's new Fable model, I was mid-task when the US government forced Anthropic to cut off foreign access to Fable 5 with little to no notice.

When Fable was cut off, I was well-prepared and had a backup plan, so it didn't have a hugely disruptive impact on the web design business and the SaaS website I started earlier this year.

We're a small team, and AI tools are a big help. I couldn't have made my website and other services as advanced as they are without AI. However, this incident was a big reminder that you absolutely need a plan B — and even a plan C — to fall back on when working with AI.

The mistake many people make is to depend on AI wholly.

When Fable came back, we were already prepping for it to disappear again

General access to Fable has been restored, and we've already been using it. This time, the first thing we did was to make sure that everything across the codebase is saved as a Claude Skill.

That way, even if we lose access to the model again, other models can reference that skill to get a good map of the codebase from the more advanced model.

Take a smarter break in your day - and see how far you get.

It's important to keep records that exist outside of the AI tool and make sure things are documented as they're developed. If Claude knows all about our code base, but it gets pulled tomorrow, would I be able to give the details over to a developer? At this stage, I think I could, because I've been documenting everything as I go. It's a fail-safe.

The key is to augment, not replace, our usual workflows with AI.

The first time Fable went down, I had a different backup plan in place

In the past, when we used Claude's Opus 4.6, it would stop mid-task because it hit the token limit so quickly. We didn't realize how token-heavy the tool was, and it left our codebase in a bit of a mess. Since we'd learned this lesson with 4.6, we were more prepared for unforeseen circumstances when we started using Fable for the first time last month.

Last month, before foreign access was cut off, I asked Fable to create a guide that both Claude and other AI models could follow. This enabled us to pass the remaining tasks to other agents when we lost access to Fable.

We passed some to Codex and others to Claude 4.8. If we hadn't been prepared this way, the Fable issue could've resulted in lots of work being out the window.

Read more from founders working with mostly AI

I hope that if this happens again, there will be better communication from the company

More notice would've been very useful, but I can understand why there was none if there's a matter of "security" at play. However, Anthropic, OpenAI, Gemini, and other AI companies, as well as businesses that rely on AI, including small companies like ours, should always have backup plans and be prepared for disruptions.

It would've been nice to have a smoother transition and not just have a model yanked offline.

Even some messaging explaining the situation within their own apps wouldn't go amiss.

Do you have a similar story to share? If so, please reach out to the reporter at aapplegate@businessinsider.com.