Should AI features be optional

So the question is, should AI be an optional feature?

  • No. I have surrendered to Skynet.

    Votes: 4 6.2%
  • Yes, but I still want some AI features. I just want to pick and choose.

    Votes: 38 58.5%
  • I refuse to use AI in any form.

    Votes: 18 27.7%
  • Don't know or don't care.

    Votes: 5 7.7%

  • Total voters
    65

winjer

Member
Almost every device released recently has some sort of AI feature. Even apps and services are changing to add AI features, regardless of being necessary or not. Or if they improve the user experience.
Most PCs, Phones, Tablets, TVs, etc, now come packed with NPUs, or GPUs that can process AI.
Microsoft is pushing Copilot in all of their applications and services. Google is doing the same with Gemini.
Apple sucks at AI, but did manage to make agreements with third parties such as OpenAI and Google.
Cars are still not fully self-drivable, but companies still pack as much AI stuff and screens as they can.

And these features can be useful at times. Though they can also make mistakes and say or do dumb stuff.
On the other hand, there are significant issues with privacy and security, with Recall being a perfect example of that.

So the question is, should AI be an optional feature? Should companies stop pushing AI so hard.
 
I voted the third option, but I feel I should explain: I don't use any AI right now because machine learning has a given error rate baked into the technology. Nothing that I would use it for is worth dealing with the error rate. If some new version of "AI" came along that didn't have the error rates of transformer based machine learning, I'd be open to using them more.

Also, 98% of the use cases the tech industry has thought of for AI just don't apply to me personally.
 
They are blocked at my job, but would be useful with some aspects of my work.

I only want to use something if I want to use it. I don't want it forced on me.
 
Third one is irrelevant at this point. It is already forced and optional in a lot of stuff, work, media, gaming, etc. and I think it's faster and more detailed than "googling". Misinformation was already out there with and without AI, except some stupid people treat its answers as universal truths. If it doesnt cut peoples jobs(it already did in some cases) than it should be ok.
 
Third one is irrelevant at this point. It is already forced and optional in a lot of stuff, work, media, gaming, etc. and I think it's faster and more detailed than "googling". Misinformation was already out there with and without AI, except some stupid people treat its answers as universal truths. If it doesnt cut peoples jobs(it already did in some cases) than it should be ok.


There's a massive step from searching the info and verifying by yourself to having a bot spoonfeeding you.

If you want a society of mindless idiots, yeah, have some bots replying all the answers, things will be great. Who needs critical thinking.
 
Last edited:
I think its fucking Bullshit that google automaticly uses AI in most searchs with zero options to disable. Like fuck me and the enviroment those pieces of shit.
 
I think its fucking Bullshit that google automaticly uses AI in most searchs with zero options to disable. Like fuck me and the enviroment those pieces of shit.
The only option (for now) is to include -noai when you search on Google, which I do almost everytime
 
Despite the markets being crazy, AI is still easily one of the most interesting things going on right now. I just don't think it should be used indiscriminately or as a excuse to backdoor computer systems or bypass data privacy.
 
On principle, I don't have muxh against AI like some people, I just use it sparingly knowing it contains errors.

The other day I used it to look for certain lyrics in songs, and it was atrocious how it got the name of the songs wrong and just outright gave wrong lyrics for song names, imagining shit.
 
AI has that fad feel I'm afraid. Like etcha sketch or something.

It's fun for stupid stuff, but when I need hard answers (which is most of the time) I scroll past the AI summary immediately because it's mostly full of shit.
 
AI has its applications. It can be useful for game graphics, or scouring manuals or documentation for information. The summary on topics can save you time.

Personally I wouldn't hold a conversation with it.
 
I'm allergic to it and I think what youtube is doing is currently the worst. What the fuck, just because my browser & OS are running in a different language than the (English) videos it doesn't mean I don't understand it and I automatically want some extremely shitty AI dubbing. I've installed a plugin for that on my desktop browser and stopped visiting youtube on mobile.

I will say it is a bit funny seeing how big corpos demand to see ANY immediate effects from the millions (billions?) they've invested in AI so far.
 
Last edited:
AI isn't completely optional. You're using it even if you don't know you're using it, especially in the things you do online. It's only going to become more difficult to avoid as time goes on.
 
I think they scrape the hell out of sites like Gaf to train shit. Specially with regards to video game marketing. You can't really scrape Reddit because lot of that shit is already AI.

I got some major dead internet vibes with EA's BF6 marketing this past week as you scanned across sites. I don't know if using AI agents for that kind of marketing is a good idea, data is very limited because it's new products every time, not an AI expert, though.

There's value in authenticity in this space because it's vanishing rapidly.
 
Last edited:
It shouldn't even be on our phones.

Not only it consumes a lot of resources but pretty much everything you type will either be stored or used for training

I think they scrape the hell out of sites like Gaf to train shit. Specially with regards to video game marketing. You can't really scrape Reddit because lot of that shit is already AI.

I got some major dead internet vibes with EA's BF6 marketing this past week as you scanned across sites. I don't know if using AI agents for that kind of marketing is a good idea, data is very limited because it's new products every time, not an AI expert, though.

There's value in authenticity in this space because it's vanishing rapidly.

Pretty much the entire Internet was scrapped to train these models. A good portion of GAF posts are probably on OpenAI Training Data and there's certainly illegal material there as well, outside of violated copyrighted works.

In fact, AI companies are looking for stuff that wasn't posted on the internet.
 
Last edited:
I definitely want to turn it all off. I don't like some AI reading all my emails to offer me a summary. Once we have self-contained AIs that exist only on our devices and have loyalty only to the user (an e-butler if you will) then maybe, but as it is, all that AI shit is just directing you towards pre-defined goals, eroding any sense of privacy, and is a precursor to MORE monthly fees, subscriptions, etc.
 
I think a lot of it is just marketing. It isn't really AI or learning, it is just some language model doing predictive text. But the buzz is AI, so you say AI and market forward.
 
AI meeting summaries in Google Meet have been so precise it isn't even funny. Had a 4 hour meeting that AI summarized perfectly into a paragraph about the issues we discussed and it genuinely impressed me.
 
I definitely want to turn it all off. I don't like some AI reading all my emails to offer me a summary. Once we have self-contained AIs that exist only on our devices and have loyalty only to the user (an e-butler if you will) then maybe, but as it is, all that AI shit is just directing you towards pre-defined goals, eroding any sense of privacy, and is a precursor to MORE monthly fees, subscriptions, etc.
The problem here is often you have to specifically opt out (if that's even possible). It should be about agreeing first to let it do its things, especially if it's violating your privacy by doing stuff like listening to a meet or scanning your text.
 
They are blocked at my job, but would be useful with some aspects of my work.

I only want to use something if I want to use it. I don't want it forced on me.
Companies blocking AI is nuts.

I use it daily at my job, all my coworkers also use it and the company even pays us ChatGPT licenses.

No matter what you do, if you have an office job, then you can improve it with AI in some way or another
 
Companies blocking AI is nuts.

I use it daily at my job, all my coworkers also use it and the company even pays us ChatGPT licenses.

No matter what you do, if you have an office job, then you can improve it with AI in some way or another
Yea its encouraged at our job but not without going through teams, you can connect to chatgpt, this way they feel like no data gets trained. I used it personally to find out whether a massive pole was out of tolerance just by uploading the spec sheet and asking it questions, this beats the regular ctrl+f. Yea everyone here uses it too, personally I don't care, sometimes they make it extremely obvious and its like bruh at least tell it not to sound like chatgpt. They also leave the — or whatever it is.
 
Companies blocking AI is nuts.

I use it daily at my job, all my coworkers also use it and the company even pays us ChatGPT licenses.

No matter what you do, if you have an office job, then you can improve it with AI in some way or another
There are things that I could use them for to help. But overall it is a security issue.
 
The problem here is often you have to specifically opt out (if that's even possible). It should be about agreeing first to let it do its things, especially if it's violating your privacy by doing stuff like listening to a meet or scanning your text.

I doubt opting out even works, otherwise OpenAI wouldn't have trained on the entire internet.

Every single security expert tells you not to share to LLMs personal or work stuff.

This is the stuff they care about:

Work/Coding stuff > Personal Identifiable Information > Medical Information > Consumer Interests > Kinks > Creative Thinking > Daily Chatter

Google is the worst and they hoard all the stuff you post, even if they don't use it.

Even when you delete Facebook, the account and some information is still there.
 
Last edited:
Top Bottom