Take my home heater/AC system for example: It's got a handful of wires that lead the thermostat. I can plug in a basic mid-20th-century thermostat to it or I can plug in a 2026 top-of-the-line network-enabled-with-AI-included device.

Tractors and cars should be like that.

Some conversations should never be done using devices that you or your organization doesn't fully understand and control.

If this means meeting in person or going to fixed locations that have secure communications channels, so be it.

If this means using only using devices that your organization can verify are secure enough to meet its needs, so be it.

The Pirate Bay: The Movie coming soon to a torrent site near you.

Look at most instruction manuals. Look at most architectural blueprints. Look at a schematic. Look at a recipe book. Look at most other "work products" that amount to human-readable instruction manuals.

Now look for the "why". "Why are we doing this in the first place." "Why did we do it this way vs. the various alternatives." Etc.

Sometimes you will see the "why" but most of the time you won't.

At least code has the advantage of having a mechanism to put the documentation near the relevant portion of the final work product.

Sounds like the "for air" version of shark skin.

And if you take out macOS as well, then what?

If it's not, it soon will be.

The energy density is 48.3âWh/kg.

Lithium-ion is better than 100 Wh/kg.

Imagine doubling the weight of your battery pack - or more. Probably not a problem for fixed-location batteries, but it's an issue if they are part of a moving vehicle.

Companies that haven't done so already will stop making new products that are impacted by viral licenses.

I'm waiting for a well-known human author to "learn the style" of AI well enough to craft a "good enough to win an award" AI-unassisted story that all the major AI-detectors flag as "very high probability this is written by an AI."

Of course it probably won't happen with any well-known author. Learning someone - or someTHING - else's style could be hard to un-learn. You don't want your future books being tainted by the "this smells like AI" stink.

Well, that would be a form of Justice.

Offer me a local or rented-tenant isolated clone of ChatGPT that is under my control, then we'll talk.

Oh, and my agents, be they human or computer, should only get "read" access, which means my financial institutions will need to provide a credentials that only have read access.

Bottom line:
* I don't trust AI not to try to make changes to my account, but I do trust my financial institutions to not allow a "read-only" login to make changes.
* I don't trust ChatGPT or the other big-name AI companies with my data any more than I have to. Maybe someday, when there are laws in place that have been tested in court, but until then, not so much.

That way I can guarantee that it won't be "connected."

ChatGPT, look at the proposed ArXiv submission and identify anything that looks like "AI slop."

I'm eleventy-ten percent sure someone will try this and twelvity-ten-percent sure it will actually work. /sarcasm

Those are two very different questions.

Once you decide that "at all" is okay, the selfishly-ideal location is "In my neighborhood" so I enjoy the tax and economic benefits, but not "in my backyard" so I don't have to deal with the drawbacks.

"Neighborhood" would be same city/county/taxing district or within reasonable commuting distance for work. "In my backyard" is close enough to be bothered by its presence.