What is AI and how to use it?

Author

Andrei-Ovidiu Ersen
May 26, 2024

Everybody is hyped about AI now.

But what exactly is AI, and how can it help you?

You may think that to build an AI you have to be an expert in mathematics and computer science and to have a very good understanding of some very abstract topics.

This is completely true. If you want to build AI from scratch.

But you can still use AI regardless of your technical knowledge. You just need to understand the basics of AI and to know which tools you can use.

What is Machine learning?

Without diving in too many details, machine learning is pretty much like human learning. AI models receive an input associated with an output. This is called training an AI model. There are many things that can go wrong in this part, but we will not go now into these details.

These AI models have a certain architecture for a certain task. Not every AI can be used for anything. An AI for image classification is different from an AI that generated audio from text.

AI architectures are inspired by the human brain, hence the name of neural network.

Now, I will give you some examples of AIs:

  1. Image classification:

These AIs do exactly what their name says, they classify images in categories. These are already used in scientific experiments, image recognition and many more. For example, recognizing handwriting can be done with AIs for image classification. Here the categories are the letters of the alphabet, and each handwritten letter is put by the AI in one of these categories.

  1. LLMs and generative AI

    The improvement of LLMs (Large Language Models) is what made AI more popular recently. Tools like ChatGPT from Open AI started the hype around AI.

Generative AIs are those AI models that generate something, like images, texts, audio. These are very different from models that just sorts data into categories. And their use in real life is much more straight forward: you generate what you want for content creation, brainstorming and learning.

Recommendation:

If you are new to AI, I would recommend you to just start using ChatGPT to teach you a skill (coding is a good time investment)

Then, you can start using AI generated images. My favorite platforms for that are:

If you want to further personalized these models in the future, you should take a look over Hugging Faces: https://huggingface.co/models. Here you can have more models for free and you can also train your own models.

Learning New skills:

Learning is now super easy with AI. Like I’ve said, ChatGPT is very useful for that. But when I wanted to use my own documents for learning, it didn’t work. Even for GPT4 it starts to “forget” after a while that it has to use my own documents. GPTs are an advancement in the right direction, but they also fail after a while, and you have to upload your documents on the OPEN AI server.

This is why I created the cheapest AI ChatBots that use your custom-made Databases. they have access to internet and to a database created by the user by just uploading .txt files and PDFs in a folder. It requires a Serper API (free account) and an Open AI API (2-3 $/month most likely). You can now skyrocket your learning, creativity and productivity with AI!

Get them here: <link>

See you in the next newsletter,

AndreiWhat is AI?

Everybody is hyped about AI now.

But what exactly is AI, and how can it help you?

You may think that to build an AI you have to be an expert in mathematics and computer science and to have a very good understanding of some very abstract topics.

This is completely true. If you want to build AI from scratch.

But you can still use AI regardless of your technical knowledge. You just need to understand the basics of AI and to know which tools you can use.

What is Machine learning?

Without diving in too many details, machine learning is pretty much like human learning. AI models receive an input associated with an output. This is called training an AI model. There are many things that can go wrong in this part, but we will not go now into these details.

These AI models have a certain architecture for a certain task. Not every AI can be used for anything. An AI for image classification is different from an AI that generated audio from text.

AI architectures are inspired by the human brain, hence the name of neural network.

Now, I will give you some examples of AIs:

  1. Image classification:

These AIs do exactly what their name says, they classify images in categories. These are already used in scientific experiments, image recognition and many more. For example, recognizing handwriting can be done with AIs for image classification. Here the categories are the letters of the alphabet, and each handwritten letter is put by the AI in one of these categories.

  1. LLMs and generative AI

    The improvement of LLMs (Large Language Models) is what made AI more popular recently. Tools like ChatGPT from Open AI started the hype around AI.

Generative AIs are those AI models that generate something, like images, texts, audio. These are very different from models that just sorts data into categories. And their use in real life is much more straight forward: you generate what you want for content creation, brainstorming and learning.

Recommendation:

If you are new to AI, I would recommend you to just start using ChatGPT to teach you a skill ( coding is a good time investment)

Then, you can start using AI generated images. My favorite platforms for that are:

If you want to further personalized these models in the future, you should take a look over Hugging Faces: https://huggingface.co/models. Here you can have more models for free and you can also train your own models.

Learning New skills:

Learning is now super easy with AI. Like I’ve said, ChatGPT is very useful for that. But when I wanted to use my own documents for learning, it didn’t work. Even for GPT4 it starts to “forget” after a while that it has to use my own documents. GPTs are an advancement in the right direction, but they also fail after a while, and you have to upload your documents on the OPEN AI server.

This is why I created the cheapest AI ChatBots that use your custom-made Databases. they have access to internet and to a database created by the user by just uploading .txt files and PDFs in a folder. It requires a Serper API (free account) and an Open AI API (2-3 $/month most likely). You can now skyrocket your learning, creativity and productivity with AI!

See you in the next newsletter,

Andrei