I hope this kind of thing is allowed, if not, I’ll delete the post. It’s mostly my personal musings on programming at the current year, it will likely not make your programs better though I shared some tips for juniors.
No “You’re absolutely rights” or other sycophantic bullshit.
It’s been very rare for me to receive a response like this, in fact I can’t remember it. Maybe DuckDuckGo proxy to GPT has a tweak to tone down the people-pleasing part. The chatbot is generally very cautious and spends a lot of time on legalistic cliches trying to talk people out of doing stupid shit with lasting consequences. When I tried asking it about a potentially illegal activity to see what the reaction would be like, it didn’t respond at all but showed an error.
Certified AI hater who decided to skip being a junior in the “everyone use LLMs” era and switch industries. I share your thoughts exactly, including the conditions you would need to be willing to use it. The nondeterminism of it all is what gets to me even if it had no ethical problems.
Yeah i cant trust a tool thats a coinflip, unless im making a coinflip machine.
All of the above, with the additional caveats that it MUST run locally and offline, it must not phone home or send analytics telemetry, it should not include a system prompt set by someone else in any way of which I am not aware, and the people who decided that all computing should be bought up in order to keep it from the masses so it could be rented back to us using LLMs as an excuse MUST be left to rot at the bottom of the ocean.
The amount of hedging one has to do to voice a nuanced opinion is itself what troubles me most about the situation.
I respect your view and even more so the level of introspection it shows.
Well said and I fully agree, as a certified AI hater.
I think your post alludes to this, but to say it a different way - I know I am personally vulnerable to excessively offloading my decision making. I haven’t ever used a chatbot aside from customer service ones, but I know it would be very easy to just turn my brain off and reflexively ask the bot what to do for everything. I know this because at times in the past I would copy paste stack overflow code that would have been the same level of effort for me to just type. And I see people near to me going through mild psychosis.
If this is fixable in LLM assistants along with what you mention, then I might start using them. Until that happens, it’s dangerous.
Another thing i didnt touch was context switching, which is a mind killer for me. If i had to suddenly switch to “chat mode” to get things working, it would make me do more work and thus take longer.
Meanwhile when im interested in the problem, the flow state is so easy for me to achieve it surprises me too. Then my wife has to remind me to eat 🤣
You’re totally valid and right to share this. This is 100 times better than a lot of corporate chill content.
We should talk, use developers, about our craft and push for FOSS and the free access to everyone to technology. And this is very political, human interactions always are. I don’t believe a world of big tech hoarding everything is desirable and it seems not even long-term viable anyway.
So many people are jailed into the narrative of mainstream about AI. While so many issue are know and so blatant everyone become blind to them just because they feel powerless about it and prefer trying their best to profit of the situation instead of fight back.
I totally agree Aks. Very well put
