Page 5

The voice on the other end acknowledges and disconnects. I don’t wait; I pull up the system architecture on my laptop and start tracing the patterns myself.

The Gromov Bratva operates on layers of redundancy. Our financial networks are built like Russian nesting dolls—surface systems for legitimate business, mid-level routing for semi-legal operations, and buried infrastructure for the things that actually matter. Each layer is encrypted and designed to look like something it’s not.

Whoever touched our systems knew this. They didn’t attack the surface. No, they went straight for the routing layer that connects our offshore nodes—the one that only exists to move money we don’t want traced.

Somehow they did it without triggering a single alarm.

I spend the next two hours reconstructing the intrusion path. The methodology is clean. Whoever did this built their approach from scratch, customizing every step to fit our specific architecture.

That takes skill. More than skill, it takes understanding. The kind of understanding that comes from studying a system until you can predict how it will respond before you make a move.

I’ve seen this kind of work before. State-level intelligence agencies operate this way. So do the best private contractors money can buy.

There’s something else here. Something that doesn’t fit.

The intrusion was surgical. The hacker mapped our routing structure, identified the nodes they wanted, and extracted data packets without disturbing the surrounding architecture. It’s the digital equivalent of picking a lock, walking through a house, and taking only what you came for without leaving a single fingerprint.

Except they didn’t take everything they could have. They had access to layers they didn’t touch. They could have done real damage. Instead, they moved like someone who was looking for something specific and had the discipline to ignore everything else.

That’s not how state actors operate. They take everything. So this was targeted.

This was human.

Kirill arrives at nine with a stack of printed logs under his arm. He’s been with me for six years, and he’s good at his job, but he’s not a strategist. He sees threats. I see patterns.

“You reviewed the report?” he asks, setting the logs on my desk.

“I reviewed your report. Now I’m reviewing the actual data.”

He doesn’t flinch. He’s used to me overriding his analysis. It’s why I keep him around—he’s competent enough to catch problems, but not arrogant enough to think his interpretation is the only one that matters.

“The pattern suggests automated intrusion,” he says. “Probably a botnet running distributed scans. The kind of thing we see every few months.”

“It’s not automated.”

Kirill pauses. “Sir?”

I pull up the reconstruction on my screen and turn it toward him. “Look at the decision points. Here, here, and here. At each junction, the intrusion could have gone multiple directions. An automated system would have followed the path of least resistance. This one chose the path that was hardest to detect.”

Kirill studies the screen. His expression shifts from confusion to something more serious.

“It chose stealth over efficiency,” I continue. “That’s a human decision. A person evaluated the options and picked the one that would leave the smallest footprint.”

“Or a sophisticated AI system.”

“No.” I shake my head. “AI doesn’t hesitate. It doesn’t weigh options based on incomplete information. This intrusion adjusted mid-process. It adapted to our countermeasures in real time. That requires judgment, not just processing power.”

Kirill absorbs this. I can see him recalculating, running through the implications.

“What do you want me to do?” he asks finally.

“Nothing.”

He looks at me. “Nothing?”

“Continue standard operations. Don’t change security protocols. If this person is watching—and they probably are—any shift in our behavior tells them we’ve noticed.”

“What about the intrusion itself?”