Rant: Sick of AI conversations

I’m not against AI completely, but yeah like you said it makes everything almost feel completely soulless.

I stopped using LinkedIn because every other post was just AI generated; Facebook is just AI slop back to back. What’s worse is people don’t realize that over extensive use of AI, especially for something even as simple as texting or just a little cognitive function, does affect you somewhat mentally. A tool is a tool - not to be used as a crutch.

2 Likes

The dash or hypen or whatever it is, is almost impossible to remove from AI.

No matter how many prompts you give chatgpt to not use the dash. It will say “Okay, I won’t” but will still let it slip in.

This struggle is a clear example that AI doesn’t have a brain of it’s own. Everything it spits out is based on the training data set.

If the data set has dashes then AI will use it.

It probably came from all the books that were fed in the training. AI can’t ignore it.


Funny thing is that a simple script that removes the dashes from the replies will always be more effective than telling AI to do it.

https://www.reddit.com/r/OpenAI/comments/1nbtl2p/wow_weve_been_burning_money_for_6_months/

The comments were not kind.

4 Likes

He deleted the post. :sweat_smile:

Why are people so afraid of backlash. Sure you were dumb enough to make AI do all the work that did not need to be done by AI in the first place. He could have setup a dedicated server running a model to do those things for less than $200, if using AI was the end goal.

More often than not I end up asking people, why do we need AI to do this when simple rule based workflow is sufficient. Now with this agentic AI crap, everyone is going even more crazy. Trying to fit it anywhere they can to show that we are using AI now.

hype raise money and so this results. if VC start saying no funding to AI, you won’t even hear a news of it from the next day.

We were discusssing a scenario sometime back with AI email assitance. One manufacturer want to sell a product to a wealthy customer and he schedules AI assitant to send persuasive mails to the customer till the deal get closed. The customer on the other hand, does not want to buy this product at all and instructs the AI assistant to send diplomatic rejection mails till further close ups are stopped.

Remaining you can conclude :slight_smile:

2 Likes

In tech communities, there is an initial rejection of methods and processes which reduce technical person’s value. Many in sysadmin forum was against cloud email first, then against ms azure like cloud services, now very negative against AI.

All these technologies already are and will be successful, will stay with us for long. Only the methods may change.

I think you’re misunderstanding the issue. The negativity isn’t against AI as a whole, but only when it’s used to replace everyday posts, conversations, and even thinking or decision-making by people who overly rely on it and don’t think for themselves. No technical person’s value is being affected here; it’s mostly self-inflicted damage, and in my opinion, it makes everything look and sound the same, which is monotonous and almost depressing to see.

That’s so true, and I remember coming across this research (Not sure, but this seems to be the right one) - https://www.mdpi.com/2075-4698/15/1/6

It found that more frequent AI use correlates with lower critical thinking scores. This needs further investigation, especially through longitudinal studies that track the same individuals over long periods of time. Still, I can only imagine how continued extensive use will lead to cognitive decline for future generations. It almost feels like living in an episode of Black Mirror.

2 Likes

Agree with your view points that AI is pushed everywhere without any discretion. However isn’t the quote above little distracting ?

Can’t AI help people with lower critical thinking augment and advance their decision making ?

Didn’t such discussions happen when computers appeared ?

That already happened with smart phones, and IIRC there was a research on that too.

Computers still were, for the most part, tools. They offered efficiencies and improvements. AI on the other hand is literally masking your deficiencies until the moment comes when you can’t use it and then the jig is up.

In my organisation I’ve seen a lot of people suddenly getting better at email writing (good for them) but then say stuff like “myself first-name” and “what we can able to do?" in meetings. I’ve also seen interns hand in extremely verbose AI-generated code and then not understand why I’m getting mad “even though the code works". I might very well be judgemental but it is true that such people are simply loading the gun and training it at their own feet. AI should be used as a tool and improve yourself in the process. Otherwise you’re only delaying the inevitable as happened to the people mentioned in earlier posts.

This book’s link has been open in my browser since almost Feb now. I had considered giving this out for the TE 20th Anniversary celebrations. I thought if nothing else I will read it myself. Guess its time to give it a read now.

https://www.amazon.in/Singularity-Nearer-When-We-Merge/dp/1847928307

I just keep wondering, what happens when we merge what companies like Boston Robotics are doing, with what Gemini AI (or other LLMs) are doing.

Imagine this guy with Chat GPT inside his head.

1 Like

Smartphones tremendously helped improving lives of billions of people too… It did more to help than harm.

Science will augment human connection with AI, it’s totally expected, for eg., neuralink like research. AI will integrate with humans and help us develop and perform better ? We are just at the initial step of such a distant future, have to evolve much more.

Better be thankful that their emails are easily readable now. They are anyway part of your org, you have to deal with them, that is our Indian educational quality at display (for eg., https://archive.is/m3etf). The escape may be more human centrist AI with the chinks ironed out !

Vibe Coding Has Turned Senior Devs Into ‘AI Babysitters’: Vibe coding has turned senior devs into ‘AI babysitters,’ but they say it’s worth it | TechCrunch
"vibe code cleanup specialist” is job in demand now. Probably, pays good…!

Unfortunate reality is that they will hold on till they shoot their feet, if they don’t improve.

TLDR; Irony

This.

I am having a migraine

1 Like

Online delivery robots (not humanoid kind though) are already working in USA cities delivering packages.

Recently there was a video where one caused firetruck to stop since it was crossing a zebra line: https://www.youtube.com/shorts/4BFrHnwkT9Y

Two confused delivery robots confused about who have the way: https://www.youtube.com/shorts/sCRTl4AvAfM

Another: https://www.youtube.com/shorts/RZRBemXNodo

Word! This is the essence of all of humanity. The answer to “who am I?” isn’t something outside. It is what one sees oneself as. One is what one “wants to be”, not what “one already is”. Learnt this very late in life.

1 Like

Discussion regarding AI reached famous teaching book “Who Am I?: The Teachings of Bhagavan Sri Ramana Maharshi”…

1 Like

Oh, didn’t know it was a book by him. Thanks for sharing.

I only knew of the 1998 Jackie chan movie of the same name:

3 Likes