Man, you won't believe the stuff that was buzzing on Hacker News today. Had to give you a quick call. Some wild things going on!

AI and Tech Drama

First off, remember Claude, the AI assistant? So, Anthropic just blocked third-party companies from using their Claude Code subscriptions. Apparently, they don't want other businesses piggybacking on their service. People in the comments were pretty hot about it, saying there are a bunch of bugs they haven't fixed, like screen flicker issues. One guy even mentioned that Google's alternative might work on free tokens if you just wanna mess around. Someone else pointed out that Anthropic's TOS says they don't train *generative* models on user data, but that doesn't stop them from using it for internal stuff like "reward models" to make the AI better. Kinda sneaky, right?

Then there was this whole thing with Cloudflare and Italy. The Cloudflare CEO was pretty vocal about Italy fining them, and the comments were wild. Some folks were asking if this means countries could lose essential internet infrastructure if they get into a spat with Cloudflare. Someone even compared Italy's law to China's "Great Firewall" and asked why Cloudflare isn't pulling out of China if they're so upset about Italy. And get this, a lot of people thought the CEO's post was super "cringy" and a typical "tech bro" move. Ouch!

Privacy and Control

Big news from Vietnam – the government there banned banking apps on rooted phones. Basically, if you've messed with your phone's software, you can't use your bank app. A lot of people in the comments were saying this isn't about security, it's all about control. They were bringing up how people are easily scammed, but others were arguing that this just makes people more reliant on their smartphones, which has its own problems like addiction. One guy even bought an iPhone SE for cheap just to use for banking, saying he barely uses it otherwise.

And speaking of control, there was a crazy story about a surveillance company called Flock Safety. Turns out, they hardcoded the password for America's surveillance infrastructure 53 times! Like, seriously? People in the comments were sharing how community groups have actually managed to get Flock contracts canceled in their cities, which is pretty cool. Shows that people can fight back against this kind of stuff.

Cool Projects and AI Breakthroughs

On the lighter side, someone did a Show HN with a memory game to teach you to play piano by ear. Super cool! The developer was really active in the comments, taking feedback and pushing updates. He even added an Ableton keyboard layout and different practice modes for showing notes. Sounds like a neat way to learn music.

And get this: a famous math problem, Erdos problem #728, was solved "more or less autonomously by AI." That's huge! People were debating in the comments about how much you can trust AI-generated proofs compared to human ones, and whether it's even "AI" if it's just a control system. Makes you wonder about the future of math and AI, right?

Open Source and Policies

Finally, the European Commission is asking for feedback on open source. They're basically trying to figure out how to best support it. The comments were full of hope that this could be a big win for open source, with people saying Linux desktop is way better than Windows now and there are tons of good open-source products out there. But some also warned about "malicious actors" trying to confuse terms and how the US might not respect EU enforcement on US products.

Anyway, that's the gist of it for today. Talk soon!