Most of today’s tech news feels like reruns. The one story worth your attention isn’t about AI at all — it’s about how governments are starting to regulate technology they don’t understand.
California Almost Broke Linux
California lawmakers tried to force operating systems to verify users’ ages. Yes, all of them. Including Linux distributions that anyone can download and modify. After massive backlash from the tech community, they’re now proposing to exempt Linux and other open-source systems.
This matters because it shows how disconnected regulation is from technical reality. You can’t age-gate an operating system that exists in thousands of versions across millions of servers worldwide. It’s like trying to put a bouncer at every library.
The real issue: as AI gets embedded into more systems, expect more regulatory overreach that misses the mark. Companies building autonomous AI systems need to prepare for laws written by people who think “the cloud” lives in the sky.
Norway Builds a Petabyte AI Training Setup
Norway’s research institutions are deploying 2 petabytes of Huawei flash storage specifically for LLM training. The setup is designed to handle the massive data throughput that modern AI training demands.
This is significant for two reasons. First, it shows how serious compute infrastructure is moving beyond the usual suspects (AWS, Google, Azure). Second, it highlights that AI training isn’t just about GPUs anymore — storage bottlenecks can kill performance just as quickly.
For companies running AI workloads, this is a preview of what enterprise infrastructure will look like. You can’t run autonomous AI teams on traditional databases and file systems. They need storage that can keep up with how fast they think.
At Kerios, we’re seeing this firsthand. When AI agents collaborate on complex business processes, they generate and consume data at rates that would break most CRM systems. That’s why we built our platform from the ground up for AI-native operations.
The Pattern
Both stories point to the same trend: existing systems — whether legal or technical — weren’t built for AI. As autonomous AI becomes normal, the gaps will get bigger.
Ready to see what AI-native operations look like? Check out Kerios.