asked to hit 'any key'????? Oh the horror!

We're going to reach a point where the tech to do this kind of thing is pretty close to invisible. It's going to be interesting when that happens.

Blue dot in red state here. Farmland is currently zoned as agricultural across most of Georgia. Two main categories exist; exclusive (use as farm only) and non-exclusive. I'm not going to search all 3 zillion counties property maps to see what the balance is but the state does use zoning to protect land.
FWIW, zoning can be challenged and changed so just because something is zoned agricultural now means nothing.
A better long-term solution is to establish a permanent conservation easement. As said elsewhere in this bit of thread, these options require the land-owner push to get the specific zoning category. I can see why a farmer would not want to restrict what their land could be used for in the event that they get old and the kids might or might not want to keep farming.
No simple answer to this problem because it's not A problem but Many problems all balled into one thing.

My spouse has felt for some time she had some type of arrhythmia. In office ECGs never detected anything. She has a Samsung watch (not the latest version) that detects abnormal heart patterns. It pops up with 'you have AFib'. Meh, figures it's bogus. During one of her episodes, she initiates an ECG, AFib is the result. At a regular doc visit she mentions it to her PCP. They open up the app, find the displayed data, and doc concurs; she has AFib. They do a 24 hr monitor and confirm AFib. In conversations with her PCP and the cardiologist the docs confirm that they're seeing people with smart watches that do, in fact, have things like AFib that are confirmed on further testing. This seems to be a real thing. I'm sure there are false positives and likely false negatives so it would be good to have actual clinical data but in the absence of real studies, if you have one of these that pops a message, see your doc.

Well, if that happens, then no, not what I want. I'll evaluate every instance of LLM assisted shopping experience. If it works, hey, happy day for me. If not, then I just don't deal with that merchant again.
I've been on the end of badly implemented LLM answering service too. Called a medical center and needed to connect to a specific department. Turns out there are two similar departments, each dealing with different body parts and the LLM directed me to the wrong one. Not just me, but everyone who called. Meh, they'll get it fixed.

That is a great question and, upon reflection, think that systems should identify as a bot and offer an option to speak to a human somewhere in the Q&A process. As I said, my order was bog standard but it's possible I could have asked for human assistance if I needed to change something from non-standard. Systems should make it clear that human help is available and how to get to that person.

FWIW, non-LLM automated answering systems (press 1 for , 2 for and none of them are what I really need) have suffered from this issue since forever and often make it impossible to reach help (I've had these systems just disconnect me if I failed to use the allowed prompt responses). So, bad customer service already exists without LLMs.

A few weeks back I called a BBQ restaurant in Vero Beach FL to order dinner. A very pleasant woman answered the phone and asked for my order. It took about 15 seconds to realize I was speaking to a bot (bonus points, it was the first day of roll-out) and I'm thinking 'here we go, gonna be another '11th floor elevator' sketch'. TL;DR, it worked perfectly for my order. Nothing I said was misinterpreted and the final order was exactly as I ordered. I spoke to another customer while I was waiting for my pickup and they did have an issue because they wanted something changed on one item (more/different sauce or something) so they hung up and just came in to place the order. If/when I order again, I may try and modify a standard order just to see what happens. I'm also curious to see if the restaurant sticks with the system.

So, for something like this, LLMs might actually work or at least work enough to make it worth while.

The ability to share data across the world improves medical care. If a local physician can't quite read a scan, the scan can be shared to instantly which improves the overall care the patient receives. That's why you want this stuff on a network.

I work in the medical device industry and none of this is all that new. It's clear that the FDA wants companies it regulates to THINK ABOUT RISK and then show that you've mitigated the relevant risks and can prove that's true. It's not rocket science, and doesn't have to be that expensive. When you build a network, you figure this stuff out anyway; all the FDA wants you to do is WRITE DOWN what you did.

It can be a pain, but it doesn't have to be any more painful that good basic application control and change management.

So, I had this friend several years back, who looked at driving safety another way. The goal is to make the driver VERY careful. His recommendation was to remove all safety gear; no seat belts, no head rests, no radar, nothing. Now, put a big metal spike right in the middle of the steering wheel and point it at the driver's solar plexus. Oh yeah, I'm not gonna hit anything EVER with that setup.

Our work email is explicitly 'for work only' and we're alerted that nothing is private.
Anything that enters or leaves our inbox, even if deleted immediately, is stored on a separate system for 10 years. It's available to us via a search plug-in to Outlook (email Xtender) so it means you don't have to keep every email you ever get, just search for it when you need it.
All outbound mail is scanned before it's sent to see if it might include anything secret (we make stuff for the military) or if it's being sent out of the US and we have to answer a prompt box if anything looks hinky to the system.
All files copied to any external drive are scanned to see if anything secret is on the way out. I haven't tried encrypting a file to see what happens.
All http traffic is scanned and logged.
Any call out of the building is logged (we have to punch in a personal code to dial out).
I'm betting the printers store data as well.

The hardware and software belongs to the company and they make sure we know it. If you're keeping anything personal on a work system, you're not using your head.

Here's the tip off... "for example marketers or small business owners"

These folks LIVE in Excel and Word. VBA is a perfectly good language that integrates with what they use all day, every day. There are tons of examples available. They typically have real problems to solve where the data exists in Excel. For that matter, LEARN Excel. That app has enough bells and whistles to keep a person busy for a lifetime; do they know how to use pivot tables? How about the offset method for chart ranges? Stats? IRR and other financial computations? Connect to remote data sets including web pages? It's all in there.

The main point, as others have mentioned, is that the key to getting into programming is to have a problem to solve, then pick a tool to solve it. Excel/Word/VBA provides an easy entry point for exactly this to happen.

If they can do VBA, they can then branch out and learn other languages.

Crap crap crap....should have read all the doom/gloom/you're gonna die if you update too soon stuff last night before I upgraded to 10.04 LTS...CRAP...

Oh wait...it's working perfectly, no problems during or after the upgrade...I'm being set up, right? WHEN OH WHEN will it all fall in the crapper and my life be RUINED?

I've found that the last couple of upgrades on both my desktop and laptop have gone swimmingly well. Hat's off to the release teams!

A few years back, the Iditarod added GPS trackers to some of the top drivers sleds so their position and speed could be displayed to users who subscribed to the 'Iditarod Insider' service. One of the guys wasn't too happy about this and gave his tracker to one of the supply aircraft...Lookie... is now going 150 mph, in the wrong direction, at 3000' agl...awesome dog team!

Actually, the experiment went over really well with those who follow the race so this year everyone got a tracker. It's pretty cool to be able to see how everyone is doing in real time. The mushers don't have access to the data so they're still going cross country using old school technology (eyes and brain).

So here's the rub. If 'the government' backs off and lets device and drug companies be less rigorous and more nimble in their work, whoo whoo...more change in less time, innovation, costs drop, new products stream onto the market...yeah.....happy day....
Until someone gets hurt because the company didn't do the things you should do when designing products that are supposed to save lives. A couple of years back the FDA did exactly this to a company making...ohhh flu vaccine. Seems they cut the company a bit too much slack and the entire batch was crap, had to be recalled, and people didn't get flu shots. Guess who spent time sitting in front of a congressional committee explaining what happened? YOU want to explain the congress why companies are killing people?
I work in the industry and as much as I chafe under the paperwork, I have to admit that the only thing the FDA is doing is making us do what we SHOULD be doing. Is it expensive? You betcha. But I, for one, don't want MY health damaged because someone wanted to get product x to market just a bit faster and for a lower cost.

Medical devices, like drugs, cost money because the company has to prove to the FDA that their device does what it says it does and that the risk the device presents is less than the benefit derived from the device. It's not uncommon for a device to take up to 5 years to get through the design/validation process. If Nintendo decides to offer the Wii balance board as a medical device, you can bet your bippy it'll cost more than $100.