AI AI AI

AI AI AI

What is AI?

Nowadays, everyone talks about AI, wants AI, and is using AI, and this is totally fine. But let’s make it clear: AI isn’t a thinking mind in a computer that is good for everything—at least not yet. When we talk about AI, we are often referring to applications like image generation, self-driving cars, drones, and, of course, ChatGPT and its rivals. Using AI mindlessly (assuming it will think for us) can lead to many problems. So what is AI, then?

I asked ChatGPT to describe what it is and how it works.

👨‍💻 me: Describe me what are you, and how does LLM works, like a stoned surfer.

🤖 ChatGPT: Duuude, alright, so check this out. I’m like this totally rad artificial intelligence, man, made by these gnarly scientists who just pumped me full of knowledge, y’know? I’m called a language model, but my bros call me LLM for short, ‘cause, like, why waste time, right?

So, picture this, dude: I’m like this massive wave of words and ideas, and I can totally ride the currents of language, man. When you ask me something, it’s like you’re throwing me a sweet surfboard, and I just shred through all these possible answers, pulling out the gnarliest ones, like, whoa.

The way I work, man, is by vibing on patterns, you feel me? I’ve seen so many sentences and conversations, it’s like I’ve surfed every wave in the ocean of language. So when you toss me a question, I’m like, “Yo, I’ve seen waves like this before,” and then I just, like, go with the flow, pulling together words and phrases that match the vibe you’re putting down.

It’s all about probability, bro. I don’t really “know” things like a dude with a brain does, but I can sense what words are likely to come next in a sentence, like predicting the next killer set rolling in. So I just keep riding that wave until I’ve formed a response that’s, like, totally tubular and makes sense in the context you’ve thrown at me.

Pretty wild, huh? It’s like endless surfing, but with words, man. So just keep throwing those boards my way, and I’ll keep shredding the linguistic waves for ya!

So these AIs were trained on vast amounts of data, including texts, images, conversations, and more. When we provide a prompt, they generate a response with the highest probability based on the given context. Even though AI isn’t a thinking mind prisoned in a computer, nor a sentient entity, it’s still impressive how relevant the responses can be.

Using AI in a good way

I use ChatGPT, Copilot, image generation tools, etc., on a daily basis since they are publicly available. Depending on how they work, they can sometimes provide very useful answers and other times offer useless or misleading information. At first, it can be difficult to determine whether the response is accurate or not.

My advice is to use AI like you would use any other tool. Don’t expect them to think for you or provide all the correct answers. If you ask something and it gives you an answer, understand what it says and how it works, and keep evaluating it until you have no doubts about its correctness.

Just like with Stack Overflow, where you shouldn’t mindlessly copy and paste the first answer, but instead examine the code snippet, understand it, modify it to fit your exact task, and then apply it.

This rule also applies when using ChatGPT or its rivals. These tools are fascinating, but they are still just tools, so use them as such.

When I use AI, I try to use it as a trigger for my own thinking, and for that it’s really good. When i stuck in something, instead of simply asking for a solution, I describe my problem and the kind of solution I have in mind. This approach can lead to more creative ideas, which are often better than just asking for an exact solution, and can help you develop your problem-solving skills.

Future of the AI

Of course, these are just speculations; nobody knows when or what will be achieved. The goal is AGI (Artificial General Intelligence), which is like the Holy Grail of AI research. Some experts say it will be achieved by the end of this decade, while others believe it will take much more time.

In my opinion (and I’m absolutely not an expert), it might not take that long and we could achieve human-level AGI in the next few years. However, the facts are as follows: research is progressing exponentially. Many labs are continuously developing their AI solutions, and there are numerous labs building robots (just think about Atlas, from the Boston Dynamics) for various uses. Additionally, many places are eager to adopt these kinds of solutions.

To be honest, it is fascinating and terrifying at the same time. I can’t even imagine how far we could go if we had an entity that could help us solve our biggest problems like global warming, poverty, and pollution. Or what technological, economic, and environmental levels we might achieve, potentially leading to a golden era or utopia. On the other hand, humanity has not always been the wisest when it comes to technology. What can be used as a weapon is often used as such, and this could lead us to an apocalypse, as many movies depict.