Spies used to chase weapons plans and state secrets. Now they want something else: AI.
CrowdStrike says China-linked hackers are going after US tech firms. The goal is to steal AI they can't build fast enough on their own.
That would help China close the tech gap with the US.
In a report out Tuesday, the security firm shared a striking number. China-linked groups were behind more than 58% of state-backed attacks on tech firms, according to CNBC.
The top target was AI. That means the models, the data, and the work behind them.
These were not random hits. They were aimed and planned.
The findings cover the year ending March 31. The hackers also kept a quiet foothold inside North American tech firms by slipping through software flaws.
The same groups went after government messages in Southeast Asia. They want both secrets and an edge.
We turn news like this into plain takeaways every morning in Market Briefs, in five minutes a day, plus a free investing masterclass when you join.
The US has blocked China from buying the best AI chips. That has slowed its progress.
China has built its own AI models to cut costs. But the best US work is still ahead, and taking it is faster than building it.
So instead of building, the cheaper path is to take. Stealing a finished model is like copying the answer key instead of sitting the test.
Code is easy to copy and hard to guard. That makes it a prime target.
This isn't only a China story. CrowdStrike said North Korea-linked workers tried to sneak onto tech payrolls to raise cash for the regime.
AI is now valuable enough to be a target for spies. That changes who has to spend to defend it.
Earlier this year, OpenAI and Anthropic both said Chinese firms pulled secrets out of their systems.
When the crown jewels are lines of code, guarding them becomes a steady cost. It also becomes a steady business for security firms.
AI models cost billions to train. A single theft can skip all of that.
Banks, hospitals, and tech firms all face the same threat. So the market for defense keeps growing.
Every breach pushes firms to spend more on defense. That spend tends to grow year after year.
Following where smart money flows in security is one way to read that shift.
The faster AI moves, the more there is worth stealing. That makes AI theft a real market disruptor this year.
The labs build the models. The hackers try to take them. The guards get paid either way.
The AI race isn't only being run in labs. It is being fought across networks too.
For more on the firms built to defend all this, sign up for Market Briefs and get a free 45-minute course on finding investments as a bonus.