Don't be sad! NetNewsWire has been my newsreader for well over a decade, and I only consider "upgrading" out of idle curiosity, because the version I have now does what I want, and does it well. The version treadmill is a machine producing only sadness.

Thanks for confirming this! I read the speculation, and traveled fairly close to Dankova. Actually climbing it is a serious land expedition from the nearest international airport.

We used to have an endless supply of new search engines, so "SEO" was not viable. Then Google got a monopoly on search, DoubleClick reverse-acquired Google, and here we are.

Become a plutocrat, or be useful to plutocrats. I don't have the moral flexibility for the former, but plutes tend to care about their images, legacies, and mewling broods. A clever person can find a way to be the latter.

"Evaluating and mitigating the growing risk of LLM-developed 0-days" would be much more interesting and useful. Try harder, guys.

> Microsoft 365 Copilot | Create, Share and Collaborate with Office and AI

That's some insight into Microsoft's brain rot, isn't it? "Imagine spending every day for the next year dealing with robot office software."

I spend most work days with robot dev software (claude code). If Copilot had a similar ability to do useful work with meaningful oversight I wouldn't mind spending time with it. Sadly it does not

That's different: you're using bots to create software that someone finds useful, so you're entirely on the producer end.

People using Copilot 365 are on both ends, both producing and being inundated with generated content. Sure, they can crank out 10x as many emails and slides, but they also have to deal with 10x as many incoming from their coworkers. They can of course use a bot for that, but then they're back where they started, but with errors.

> consumers realising the limits of "AI",

This is my guess for the demand side: most people will drift away as the novelty wears off and they don't find it useful in their daily lives. It's more a "fading point" than a "breaking point."

From the investment/speculation side: something will go dramatically against the narrative. OpenAI's attempted "liquidity event" of an IPO looks like WeWork as investors get a look at the numbers, Oracle implodes in a mountain of debt, NVidia cuts back on vendor financing and some major public players (e.g. Coreweave) die in a fire. This one will be a "breaking point."

> bespoke little programs that do one thing and can be discarded after use.

That's my best-case scenario as well: LLMs are scripting languages for a broader audience. They just barely automate busywork, but are not a reliable foundation.

I had the option to get a "Real ID" the last time I renewed my driver's license, and did not. I forget which stupid bit of paper gave me trouble, but I had a valid passport (the Mother of All IDs), which was both insufficient to get a "Real ID" and sufficient to fly. It's a joke, a nuisance, and now a revenue source.

You're not going to believe it, but if you already have a passport - you don't need Real ID (in ideal world). I only got Real ID because I want to have zero questions about my immigration status.

>You're not going to believe it, but if you already have a passport - you don't need Real ID (in ideal world). I only got Real ID because I want to have zero questions about my immigration status.

DHS says they don't consider RealID to be reliable "proof" of immigration status[0][1], so you might consider rethinking that strategy.

[0] https://www.biometricupdate.com/202601/dhs-agent-tells-court...

[1] https://storage.courtlistener.com/recap/gov.uscourts.alsd.76...

> DHS says they don't consider RealID to be reliable "proof" of immigration status[0][1], so you might consider rethinking that strategy.

I did, I became American and now have a US passport. That is just moving of the goal post by current admin.

If accurate, this strikes me as something like malicious compliance.

> Translations, which help you browse the web in your preferred language.

Machine translation can be useful when you want to get the gist of something in a language you don't know.

> Alt text in PDFs, which add accessibility descriptions to images in PDF pages.

OCR? Okay...

> AI-enhanced tab grouping, which suggests related tabs and group names.

What is this feature even trying to do? It sounds like ill-defined trash.

> Link previews, which show key points before you open a link.

Or I could just click the link.

> AI chatbot in the sidebar, which lets you use your chosen chatbot as you browse, including options like Anthropic Claude, ChatGPT, Microsoft Copilot, Google Gemini and Le Chat Mistral.

This is the thing that most people are probably complaining about. Lumping the other features in with it is a distraction.

> The AI features can be disabled entirely or individually, so users can pick and choose what they want to use

It sounds like you would want to switch off two of them and leave two of them on, no? How is that malicious compliance?

The master AI switch is for people that have moral issues with all AI, so they want all future features turned off.

Mozilla is grouping a bunch of unrelated stuff in with the one thing people don't want.

That's because "AI" is a bunch of unrelated stuff that happens to use LLMs. Maybe you don't agree that machine translation using a large language model is AI, but other people do.