Problem is micromanaging executives that are all in and demanding to see some volume of LLM usage the way they think is correct (little prompt, large amounts of code).
Thus practice may be very bad for your health. Not that these "executives" care, but you should.
Yes. And that is how AWS got their 13 hour (?) outage. That outage was probably more expensive than what they can save in cost over a year or several by using LLMs as surrogate coders.
Why stop there? Make it part of the start-up message and if there is none, add one!
Well, the routinely clueless economics graduates certainly think so. My take is that in a few years actually competent coders will be in high demand to fix the mess and out out a lot of fires. When that happens and if you are inclined to participate, make them pay through the nose.
Sounds like very risky behavior. But some people are simply not smart...
Will probably not take long.
Wouldn't building GPS into vapes make them much more expensive, especially for something that doesn't work inside buildings? That being said, Google does a fairly good job of triangulating your position from nearby WiFi access points, but WiFi would be even more expensive than GPS.
TFS says the vape would use Bluetooth to connect with the phone/app. The vape won't work w/o the app, so they'd build any extra functionality into the app, like ads, tracking and geo-fencing. See my other post Why stop there?
Move the vape too far away from the phone, and it shuts off again.
If the vape is going to be app-locked, then the app can be (a) tracked and (b) geo-fenced. Establishments/locations can post if they allow vaping and the (nanny) app can track where you are and disable the vape if it's prohibited. Marketing bonus: The app will probably be running a lot, so the data it could collect could be HUGE.
At least with AI you can make it document the code it writes and the architecture.
Who was it who said that no documentation was better than incorrect documentation? Aside from you in the not-too-distant future, I mean.
Seriously, this is right up there with silly nonsense like "just have AI write the tests as well". It's like you hate your future self and want them to suffer.
Fortunately you can also tell it to evaluate and document the legacy code.
Yes, I suppose you can tell it to do that
Very helpful.
Is it really? Every AI fluffing article we've seen that makes those claims has, after even cursory examination, turned out to be complete horseshit. I suspect that, like most people who still have faith in magical AI, you're not paying too much attention to the output.
Your mode of life will be changed to EBCDIC.
