Man..
Every day feels like AI is being pushed so far and for the wrong reasons.
From being used to farm user data for the sake of publicity that its not even right to "sophisticated" scams to lifeless, soulless AI slop that tries to imitate art. Like, damn, Twitter users now have a way of flooding your comments section using AI scam bots trying to steal your crypto monkeys! Nothing good is coming from these state-of-the-art LLMs.. At least that's what I hear from all the news on AI nowadays and maybe not only me but the entire internet as well. They despise AI.
BUT
People are hating AI but not disregarding its potential. Huuuuge steps have been made in different areas thanks to AI. We may have a better way of detecting breast cancer, teachers can search for more or better material for their classes, help people with disabilities work on papers by correcting their work, learn new languages, etc.
My theory
The use of AI is really on its way of changing millions of lifes for the better, we are just not able to see it because, and this is a theory of mine, it is not being digested in a way that attracts curiosity.
Everyone is curious about the use of AI, its today and tomorrow's hot topic but how many people do you know that actually go into a research paper or one of those AI blogs, even with all the bloat it has, and reads their articles/findings? If it wasn't because they have to do it either for homework or work-related, they wouldn't have known about it.
How to fix this?
What I'd propose is fight against these commonly heard news of the latest AI scandal, spread the misuse of AI and also share your latest favorite AI development to these platforms. I'm not telling you to start copy and pasting the usual 🧵 news guru 🧵 with the thread emoji on his post. Instead, people would like to see actual interest put into these posts. I know I would and it would really help in this gray looking cloud that is the AI world.
I am curious though.. How would you fix this? Do you agree? Are there any good AI news portal that you'd recommend? I'm really interesting in seeing your comments. All of this was something I really needed to get out of my chest :)
Top comments (11)
I like your post, but I see some issues with it. You are saying This is why, models can never be 100% accurate? How can you know this, we cannot tell the future so saying something like this is false and discredits your credibility in my opinion.
We understand the world around us through our senses, which take in input. There are many different input devices in the field of engineering.
The new models are getting better at being omnidirectional, taking in the world around them.
If we could invent the transformer to “just predict words,” we can probably invent something even more advanced. With the help of AI that can amplify our learning and understanding, new achievements can boost the development of innovative technologies. I think you are underestimating this a lot. I have never seen any kind of technology advance at the same speed as these new systems. They have been around for a while, but what’s happened in the last year is, in my opinion, crazy. When someone says it can’t do X or Y and will never be able to do this or that, it has proven naysayers wrong over and over again.
I think your static mindset about the capabilities of these systems is short-sighted.
Just my opinion—sorry if I was harsh, but I’m tired of the skepticism when we see results that are frankly unbelievable.
I get it and now that I think about it, it kinda contradicts with what I've said in the post. There's been some huge advances with AI in many fields, we're only some searches away from knowing that and they definitely should not be diminished or just taken lightly. While I was writing all of that down, I only had ChatGPT and the other text generative models in mind and it was pretty short-sighted of my part.
Thanks for pointing that out!
I was thinking the same thing! Their "world" is very veeeery convoluted because there isn't a single answer to what even the world is. I think, the fact that information, nowadays at least, can't be found in a precise way like it used to be before really hurt this LLMs during training and now we can see it during deployment.
And what about the billions, that have no access to this technology at all?
I don't have the answer to that question. That answer is in the hands of their governors and if they really want to do good with the current technology, sadly.
My greates concern about AI is not what AI does, but what the people will (or can) do with it.
You and me both, man. It's really really concerning..
I feel you bro! I myself have a lot of feelings regarding Ai. I use it all the time and it has helped me tremendously, but im a bit scared of the future..
I am too. AI is covering too much too quickly.