I travel a bunch! The real-time traffic re-routes have helped me more times than I can count! I second guessed them at first, but I have come to trust them at this point.
1 Like
So it’s a Google Maps topic now, apologies to our beloved agents! 
It would change the route if you didn’t do anything*, and it’s a real bummer. Specifically: the inability to disable this feature. I’ve made my decision, the program should not change it on its own! Even if you notice the on-screen notification in time, tapping the phone at that exact moment can be both difficult and dangerous. Navigation software shouldn’t cause stress or distraction.
*) I’m talking about what I’ve experienced; the app and its behaviour are constantly changing
I’m with you on that one. I do seem to recall feeling a bit rushed to make up my mind whether to take the suggestion or not. But I’ve always managed to tap yes or no in time… so far.
1 Like
Can we says its swerving, back and forth, about how the assistants are trying to help, often succeeding, and often not, improving as there is feedback, some regulatory, some by market forces, and some by emerging capabilities we never envisaged or were cobbled together in a novel format?
With a bit of nostalgia and hindsight thrown in.
-Dick Tracy wristwatch? My grandfather spinning in his grave.
-Is that with HomeAssistant Companion installed on it sir? My father spinning in his grave?
-I’ve seen WWWIWYS (what we want is what you see), but WYWIWYG (what you want is what you get) isn’t quite there yet.
What novelties await?
1 Like
it great great that “WYWIWYG” . 
Looking back, almost every generation had its own “that will never happen” moment. Then, twenty years later, it becomes so ordinary that nobody talks about it anymore.
Maybe that’s the real pattern we’re all describing here.
As for what comes next… I have no idea. That’s probably the fun part.
I do see @NathanCu 's point though. And share it to a large extent I’m finding. I told o colleague the other day, who asked me what AI actually was. I told him: “look at the table in front of us and imagine a hammer laying on it. Neither of us have ever seen one, no clue what it’s for. But we’d soon figure out the thing has a handle, is heavy, can smash things, etc.. We could work out quickly that it is tool of some sort and largely what it’s for. AI is another tool, but a much more abstract one. Not self-explanatory at all. It is so diverse you could do pretty much anything with it. It is a tool like any other we do know. And like those we know, we have to learn exactly how to use it (proper prompting for example). Otherwise you’re just trying to drive a nail with the handle”.
I could literally see that sink in, then his eyes widened. And ideas started flowing. Told him to stop being a monkey wanting to smash things with a hammer’s handle and to first learn to wield the hammer properly.
And this scales up perfecty. Gonna have to have a a similar convo with senior management of the co. I work for soon, and I’m pretty sure I’ll be telling them the same thing. They have a similar position NathanCu was describing.
Edit:
To be clear & honest, I still consider myself somewhat of a monkey in that regard.
1 Like
THIS!!!
So much this…
(I can enjoy coffee today someone listened.)
My conclusion’s gonna be: But hey, I’m just a HVAC field service tech. What do I know.
Anyway, I’m on holiday. So, back to work improving Greg and @RedKing 's putting in some miles, gotta keep up lol
1 Like
Coffee should be enjoyed, no matter what.
1 Like
Spot on. As millennials who remember the early days of endless web tinkering, it’s frustrating how much autonomy we traded away for ‘convenience’ in the cloud era.
I totally agree with the hybrid outlook—we don’t need to throw the baby out with the bathwater, but having local models at home for our smart spaces is a game-changer. It brings back the pure joy of building something with your own hands, configuring things just the way you like, and knowing your data actually stays yours.
3 Likes
it’s good idea hammer analogy. It actually reminds me of the early days of internet.
Back then, many people looked at a computer and asked: “What can I do with this thing?” But gradually people discovered that the computer itself was not the point. The interesting part was what people could create with it.
I think AI is similar, but maybe even harder to understand because the “hammer” is invisible. A normal tool has a clear shape and purpose, but AI is more like a toolbox where nobody has discovered all the tools inside yet.
And yes, many of us are probably still monkeys holding the wrong end of the hammer. 
We are all learning how to use it.
The exciting part is that every generation has gone through this process. First we learn the tool, then we build new habits around it, and eventually the tool becomes so normal that the next generation cannot imagine life without it.
1 Like
I also agree with your point about not throwing the baby out with the bathwater.
Looking back, every generation seems to trade something for convenience. The early internet gave us freedom to explore, but later platforms made everything easier while taking more control into centralized services.
Maybe the future is not choosing between convenience and control. The interesting challenge is finding a balance — where AI can help ordinary people without turning their homes and data into someone else’s platform.
That’s actually why I’m interested in local AI and open systems. Not because cloud AI is bad, but because people should still have a choice.
I think most people actually believe AI will help humans work and live more efficiently, rather than simply taking everyone’s jobs away.
We have seen this many times before. The introduction of ATMs did not eliminate bank tellers as many predicted. The internet changed newspapers dramatically, but people still work in journalism and media today. The nature of the work changed.
Sometimes people look at a smart home example and say: “Why spend thousands just to turn on a light?” But I think they are missing the point. That light is not the whole story. It represents an exploration of a new interaction model, and some of those experiments will eventually lead somewhere unexpected.
Maybe the question is not “What should AI do?” but “How do we learn to use this new tool?”
Going back to the hammer analogy, we are probably all still monkeys holding a hammer for the first time. We may even spend some time trying to hit nails with the handle before we figure out the right way to use it.
That learning process is probably part of every technological revolution.
1 Like
Still is the best to do, to avoid “Touristic Clouds” and commerce, or simply to get a nice relaxed time in New Towns
1 Like
Exactly. Learning new tools also means knowing when NOT to use them.
1 Like
Some new tools are more complex than a hammer, so you spend time learning to use them, during which you also learn a lot about the job in hand.
Then suddenly you think… “Hang on. I could have done this with a hammer.” 
AI is tricky in this respect - you can learn masses trying to make it work in your particular scenario, but the temptation is not to, and maybe to overlook the fact that you don’t need to.
I was at Woodstock, you know. Now you dump your crockery on me. Sorry, that’s Greg. He’s got the open mic bug and now I can’t shut him up.
2 Likes
Also true, that.. But even then you’ve effectively learned 2 things: the ai way + everything that comes with it, but also how to stick (pun intended) to the hammer next time for that specific task. So never really a waste.
In my case, I quickly found out you have to keep your eye on what ai is doing for you, even while it’s doing the work for you. Much like airline pilots. Planes can mostly fly themselves for years now, but for obvious safety reasons they are managing everything from start to finish. Just in case they need to make any adjustnents or take over altogether. Anthropic found this out the hard way.
About Greg, that’s hilarious! But i did build a swich for that, anything i need to fix? Do let me know, ok?
1 Like
No, all is well. I just say “Hey Jarvis, tell Greg to shut up.”
2 Likes
That might be my favorite use case for AI so far.
“Hey Jarvis, tell Greg to shut up.”
I suspect Greg will eventually find a way to ignore Jarvis too. aha! ALL IS WELL.
1 Like
The pilot analogy is a really good one.
I think AI is at its best when it feels like a good co-pilot, not when it quietly flies somewhere you never asked to go.
1 Like