Something extraordinary has just happened in artificial intelligence, and it deserves more attention than received.
Anthropic recently announced its system of choice it’s not to release openly. Model – Claude Mythos previewIt is reportedly so effective at finding and exploiting software vulnerabilities that the company restricts access to a small group of organizations responsible for critical Internet infrastructure.
That alone should give us pause.
Red team
In cyber security, there is a practice called “red teaming” – thinking like an attacker to expose vulnerabilities. It is a demanding craft creativitydeep technical knowledge and the same opponent imagination. The best practitioners can uncover subtle flaws hidden within highly complex systems.
Anthropic claims that this AI can outperform almost all of them. Just not faster. Just not cheaper. Better.
The model reportedly uncovered critical vulnerabilities in widely used systems, including ones that had gone unnoticed for decades. In one case, it discovered a flaw in software that had long been considered one of the most secure programs in the world. In another, he discovered a bug embedded in code that had been executed millions of times without triggering alarms.
What is surprising is not only the discoveries, but also how they were discovered.
These are not linear problems. They require connecting distant dots, integrating multiple small weaknesses into a coherent path, and exploring possibilities beyond conventional thinking; they require navigating such a vast search space that even experienced people only explore a small part of it.
Here comes something new.
At the dawn of a new era
We’re starting to see systems that just aren’t there to help human thinking but operating beyond its natural limits—systems that can explore more possibilities, test more hypotheses, and uncover solutions that wouldn’t occur to us. And cyber security is only the first domain.
Consider logistics – the global choreography of goods, infrastructure and information. Under normal conditions it works well enough. But in moments of disruption—a natural disaster, a geopolitical shock, a sudden surge in demand—it becomes a tangled, fragile system. Decisions are made with incomplete information, coordination breaks down, and small problems become big problems.
Now imagine applying this new type of AI to this system.
Intelligence that can scan the entire network in real-time, simulate thousands of potential interventions, and detect inadvertent actions that stabilize the whole. Not by optimizing one part, but by identifying hidden leverage points throughout the system. The same pattern applies to health care, climate response, financial systems—any area where complexity exceeds human cognitive limits. And this is a deeper shift.
For most of history, expertise has meant seeing something that others cannot. But what happens when there are patterns that no one can see—not because we lack intelligence, but because the problem space is so large?
This is Kensho
In Zen Buddhism, there is a concept called kensho – a sudden look at the true nature of reality. Not complete enlightenment, but a flash of clarity that changes how you see everything that follows. It’s short, partial, but sure. That’s what this moment in AI feels like.
It is not completely artificial yet common sense. It’s not a catch-all system. But he is is considered something new: a clear and visible intelligence beyond human capacity in certain domains, especially those defined by complexity, ambiguity, and hidden structure. And if we get it right, it doesn’t diminish us—it expands what it means to be human.
Because the real promise here is not that machines will replace human judgment, but that they will expand the limits of what humans can perceive and coordinate. A world where supply chains don’t collapse stress but real-time adaptation. Where diseases spread faster because no doctor can see patterns. Climate responses are predetermined, not reactive.
Intelligence becomes a common resource in this world, no less.
And the measure of progress is whether we are able to see, decide and build together. This may be the first glimpse of a person’s inner intelligence. And like all kenshō moments, it offers a choice: to ignore it, or to allow it to change our way forward.
