This is The Stepback, a weekly newsletter breaking down one essential story from the tech world. For more about AI's impact on the web, follow Emma Roth. The Stepback arrives in our subscribers' inboxes on Sunday at 8AM ET. Opt in for The Stepback here.How it startedThe digital publishing industry has relied on Google for a very long time. Many outlets spent time and money optimizing their webpages to show up at the top of the search engine, with the goal of getting readers to click on their articles or discover their websites for the first time. For years, the relationship between Google and publishers was largely reciprocal; online pub …Read the full story at The Verge.

For almost the entirety of 2026, I've run countless experiments with several dozen local LLMs week after week, and yet it feels as though I've barely scratched the surface of how capable local AI has become in such little time.

Long before we had the ChatGPT vs. Claude vs. Gemini vs. Perplexity debate, and even before people started arguing about Spotify vs. Apple Music, there was another debate that had tech enthusiasts divided: which web browser is the best?

BBC's Blue Planet III is looking to prove its authenticity by leaving in the less-than-perfect shots. Previous incarnations of the nature series were defined by their gorgeous, crystal-clear footage that seemed almost too good to be real. But one dead giveaway of AI is when things look too polished and seamless. To counter that, Blue Planet III goes warts and all.Executive Producer Elizabeth White told The Guardian in the past, "the amount of plankton and things" floating in the water would have caused them to cut a shot. Or "we might have framed out a boat or a buoy or gone: 'Oh, actually … the water conditions are not really good enough …Read the full story at The Verge.

Running a local LLM and a game on the same PC usually means one of them will have a bad time. A typical gaming GPU with 16GB of VRAM might have enough memory for a decent model or a game, but running both at the same time? That's a pretty tall order, and it'll likely cause data to spill over into your slower system RAM.

CrystalDiskInfo has been synonymous with SSD and HDD monitoring for years. It reads your drive's Self-Monitoring, Analysis, and Reporting Technology (S.M.A.R.T.) attributes and presents them on a single screen for easy reference. Checking your SSD's health percentage, temperature, or other metrics is pretty simple with this free tool.

Strix Halo has spent nearly two years answering a question Nvidia wasn't: how do you run big local AI models on a Windows laptop without a desktop GPU? AMD's Ryzen AI Max+ 395 pairs a strong integrated GPU with up to 128GB of unified memory, enough for a Strix Halo mini PC to generate things locally, and the Asus ROG Flow Z13 has been my portable local AI machine. Then Nvidia launched RTX Spark, with plenty of Blackwell GPU cores and a surprise; Windows on Arm for the OS.

People have been doing some really cool things lately by pointing their webcams at their screens. Just a few days ago, we saw someone breach their Samsung TV by letting Claude take photos of the screen while it was working. Now, someone has used Claude to create a light gun-esque digital shooting range that uses a red laser instead.

Some European countries are taking their first steps toward digital sovereignty by removing their dependency on foreign tech. One such hurdle is getting rid of Microsoft products, such as Windows and Office, which is easy enough on paper. The only question is, what do you replace it with?

The TV is one of the most important devices in any home, as it's the central piece of an entertainment hub. Having a high-quality panel and sound system is crucial to a satisfying home theater experience, but using your TV's default settings can easily ruin your experience. It still boggles my mind how the average consumer is expected to know which settings to disable on a new TV before they can enjoy it to the fullest.

Claude Pro proves its worth most of the time. Anthropic's models are accessible pretty much anywhere through API, but the Claude workspace has become almost necessary in my workflow, especially Projects and Artifacts - and more recently, Claude Docs too. But when I scroll back through weeks of chats, I realize the sessions themselves are pretty basic, they mostly revolve around helping me understand things. So ultimately, I'm paying for the desktop app features, but these can be replaced by local tools like AnythingLLM or even PewDiePie's Odysseus.

Lately, I've been thinking a lot about the Steam Machine, and what its actual value proposition is to gamers like myself. Underneath Valve's neat little cubic form-factor, its GPU delivers performance close to a discrete RTX 3060, but that's a GPU tier that has been available in the market since 2021.