U.S. gov bans Chinese AI

This link leads to one of the most interesting things I’ve seen in AI recently.

What’s cookin’? The US government has ordered civilian agencies to remove AI tools from Chinese startup DeepSeek from their devices, citing concerns that sensitive government data could end up on servers in China. The ban goes beyond DeepSeek’s own chatbot, covering third-party tools built using DeepSeek models too. Officials say Chinese law gives the government broad authority to access private company infrastructure, creating a potential security risk when government workers enter sensitive information. DeepSeek became a major AI player after releasing surprisingly capable, low-cost models beginning in 2025, and recently launched its newer V4 family.

🤔 Hunter’s take: The interesting part here is the third-party rule. You might never open DeepSeek yourself and still use software built on top of it. That makes AI security a supply-chain problem. Organizations increasingly need to know not just which AI tool they’re using, but which models, servers, and companies are quietly sitting underneath it.

What’s cookin’? Researchers at Penn Medicine developed an AI model called Percival that can pull much more information from a CT scan than doctors were originally looking for. Trained on roughly 400,000 scans from about 50,000 patients, Percival looks for subtle patterns tied to diseases and future health risks. In testing, it detected signs linked to conditions like atrial fibrillation, chronic kidney disease, pancreatic cancer, cirrhosis, and pneumonia. The big difference is that it does not need to be told exactly what disease to search for. It can analyze the scan more broadly and flag patterns that might otherwise be missed.

🤔 Hunter’s take: This could quietly change what a medical scan is worth. A CT ordered for one problem might also catch another disease before symptoms appear. That turns imaging from a one-question test into something closer to a broader health check. The upside is huge, but so is the responsibility to make sure those extra findings are accurate and actually useful.

Here’s your chance to get a straightforward answer, from an expert in the field.

Hi — I’m Hunter, a PhD candidate whose work has appeared in major academic journals and popular tech outlets. I founded FryAI to make staying ahead of AI clear, accessible, and fun.