The Anthropic Fable 5 and Mythos Shutdown Is Not the Story. This Is.

¡AY AY AY, AI!

Weekly newsletter to help you master the new rules for success in the AI age

This week, a few friends texted me about the Fable news. A few more forwarded articles asking what it meant and whether they should be worried. All of them smart, capable people who follow AI closely. None of them quite sure what to make of this one.

I get it. Here is what happened.

On June 4, Anthropic published a report titled "When AI Builds Itself," calling for a globally coordinated slowdown in frontier AI development. The headline finding: Claude now authors more than 80% of the code merged into Anthropic's own codebase, and engineers are merging eight times more code per day than they were in 2024. Five days later, on June 9, Anthropic released Claude Fable 5, its most powerful model ever made publicly available, a version of the Mythos system that had been restricted since April due to its advanced cybersecurity capabilities. The timing was not lost on anyone: the safest option and the competitive reality do not always point in the same direction.

Then came Friday.

At 5:21 pm ET on June 13, Anthropic received a government directive ordering it to immediately disable Claude Fable 5 and Claude Mythos 5, citing national security concerns. Behind the scenes, Amazon had called administration officials the night before with a report claiming they were able to jailbreak the Mythos model in ways that could pose a national security threat. The government gave Anthropic 90 minutes to comply. Rather than scramble to build nationality-based restrictions in that window, Anthropic pulled both models for everyone.

The order, signed by Commerce Secretary Howard Lutnick and sent to CEO Dario Amodei, demanded suspension of access for any foreign national, whether inside or outside the United States, including Anthropic's own foreign-national employees.

The debate over whether this was warranted is still live. Anthropic told the administration that the alleged jailbreak was relatively simple, could be achieved with other models, and did not demonstrate a flaw in Fable 5's safety systems. Luta Security CEO Katie Moussouris, who reviewed the Amazon report at Anthropic's request, said the government's response "seems way out of line with what's actually in the research report." Others say the caution is exactly right given what these models can do. Both sides make real points.

What is not in dispute: every other Anthropic model is completely unaffected. Claude Opus, Sonnet, and Haiku are all available and working normally. Fable 5 and Mythos 5 will return once security measures are reviewed and adjusted. The train has not stopped. It has briefly changed tracks.

Here is what I told every single person who reached out: paralysis is not protection. The news cycle is not going to slow down so you can catch up. And fear, on its own, does not give you any more ability to navigate what comes next.

The only way through is to understand what you are looking at well enough to lead through it. Not ahead of everyone. Just enough ahead of your own uncertainty to keep moving with intention.

That is exactly why Nikki and I built the FlipWork™ AI Reinvention Sprint. Eight weeks of structured reinvention with a cohort of like-minded leaders, designed to help you move through moments like this one with confidence and clarity instead of waiting for the uncertainty to pass. Because it will not pass. It will only move faster.

Caminante, no hay camino, se hace camino al andar. Wayfarer, there is no road. The road is made by walking.

My mother kept this on her classroom wall for thirty years. She taught it to every student who stood at the edge of something new and did not know where to put their foot first. She was not telling them the path was easy. She was telling them the path does not exist until you take the first step.

That is where we all are right now. Every one of us. Including the people building the technology. We are pioneering this together, and the leaders who will emerge from this moment with their judgment intact are not the ones who waited for the dust to settle. They are the ones who chose to walk.

THIS WEEK'S INSIGHTS

The fear is real. Here is what the data actually says.

The story is not that AI is taking your seat. It is that your seat is getting more important.

Microsoft's 2026 Work Trend Index, drawing on 20,000 AI users across 10 countries, made a finding that cuts directly against the dominant fear narrative. As agents take on more execution, 58% of AI users say they are already producing work they could not have done a year ago. When workers were asked which human skills matter most as AI takes on more, they chose quality control of AI output first and critical thinking second. 86% say they treat AI output as a starting point and stay responsible for the thinking.

But here is the gap that matters for leaders. Only 19% of AI users are in what Microsoft calls the Frontier zone, where both organizational capability and individual readiness are high and reinforcing each other. In a week where the most powerful AI model ever released to the public was pulled by government order three days after launch, sitting in the 81% is not a neutral position. The pace is not slowing. The stakes for human judgment are rising.

1 MYTH TO REFRAME

Myth: "When things are uncertain, the right move is to wait and see."

Why we believe it: Caution feels like strategy. Waiting feels like protecting yourself from making the wrong call.

Reframe: Waiting is a decision. And it is not a neutral one. In a week like this one, the professionals who have done the structured work of reinvention know how to read what is happening and keep moving. The ones who waited are now operating with less time, less fluency, and less confidence than before. Caution without preparation does not reduce your risk. It just delays your exposure to it.

What to do: Identify one AI-related decision you have been postponing. Not "should I learn everything." Something specific. A tool to try. A workflow to redesign. A conversation to start with your team. Start there. This week.

TRY IT THIS WEEK (Micro Actions)

Three moves you can make.

1. Forward this to someone who asks you about the Fable news this week. The person who reaches out, confused or worried, is the person who is ready to move. They just need a direction. Send them this. Tell them that Thursday is four days away and that there is still a seat in the sprint.

2. Write down what you are waiting for. Open a blank note and write the one thing that is keeping you from committing to your AI reinvention. Not a list. Just one thing. Then ask yourself: will that thing be resolved in three months, or will there be a new version of it? That answer tells you everything.

3. Take the FlipFactor™ diagnostic. Eight minutes. A real read on your Agentic Velocity and the specific edges you need to build, not a generic result, but a starting point calibrated to how you actually work. Start here.

POWER TIP

The sprint starts Thursday. Here is what one member of our first cohort said after completing it:

"What I came out with at the end was literally game changing both personally and professionally. Most incredible looking back was how part of the curriculum involved standing in our reverence as a human being and what we're going to contribute to the world from that place."

A founder with decades of experience and a proven track record said the most incredible part was not the workflows. It was the human piece.

The FlipWork™ AI Reinvention Sprint is an eight-week live program. Sessions run Thursdays at 3 pm CT and are recorded if your schedule needs flexibility. Between sessions, you have around-the-clock access to the FlipWork™OS platform.

Enrollment closes before Thursday.

Enroll now

THE SPRINT STARTS JUNE 18

The FlipWork™ AI Reinvention Sprint is an eight-week live program. Sessions run Thursdays at 3pm CT and are recorded if your schedule needs flexibility. Between sessions, you have around-the-clock access to the FlipWork™OS platform.

This is not a passive course. It is structured reinvention with accountability, real feedback, and frameworks you build and use immediately. Enrollment closes as we approach the start date.