Another day, another stack of AI headlines. Here's what actually mattered today and what I think about it.
OpenAI's Lawsuit Pile Just Keeps Growing
OpenAI is now staring down more than 50 lawsuits over alleged harm from ChatGPT. Thirty of the newest complaints come from survivors connected to a Canadian school shooting. That is a heavy number and a heavy topic, and it is not something to breeze past.
Why it matters: for years the big AI labs got to ship first and sort out the consequences later. That era is ending fast. Courts and plaintiffs are catching up to the technology, and the volume of cases alone tells you this is not a one-off PR problem anymore.
Robert's take: I have said it before, growth at all costs has a bill that comes due eventually. I am not going to pretend to know how any single case shakes out. But when the lawsuit count hits triple digits territory, it stops being noise and starts being a real signal that safety work has not kept pace with deployment. Companies building these products need to take that seriously, not just legally, but morally.
Meta Finally Locks Down the Sneaky Recording Problem
Meta is updating its AI glasses so they actually stop recording when someone covers the capture LED. The update also cracks down on covert filming and harassment uses.
Why it matters: smart glasses with cameras built into the frame were always going to run into this problem. People do not love being filmed without knowing it, and a tiny light that is easy to tape over was never a serious safeguard.
Robert's take: this is a fix that should have shipped on day one. Good on Meta for doing it now instead of waiting for a scandal to force their hand, but let's be honest, plenty of folks predicted this exact issue the second these glasses hit the market. Building trust with wearable cameras means the hardware has to make sneaky recording hard, not just against the rules.
Google Wants Alt+Space to Be Your Gemini Button
Google rolled out a new Windows app that turns the Alt+Space shortcut into a dedicated launcher for Gemini, making it quicker to summon than opening a browser tab.
Why it matters: this is a straight up land grab for keyboard real estate on Windows machines, and it puts Gemini in direct competition with Copilot, which Microsoft has already baked into the OS.
Robert's take: I get why every AI company wants a permanent seat on your keyboard, but my desktop is turning into a pileup of assistant shortcuts fighting for the same real estate. At some point users are going to want one default, not five companies racing to claim Alt+Space, Windows key, and everything in between.
The US Calls Out Six Chinese AI Companies for Copying American Models
The US government named six Chinese AI companies it says copied American AI models rather than building their own from scratch.
Why it matters: this is the latest move in the ongoing US-China AI competition, and naming specific companies publicly is a louder step than the usual vague warnings about intellectual property.
Robert's take: nobody should be shocked that copying happens in a race this fast and this lucrative. The real question is whether naming names actually changes behavior or just makes headlines. I will believe it matters when there are real consequences attached, not just a list.
That is the roundup for today. More tomorrow.