Understanding AI Basics and How It Works

Transcript

Hello everyone. Welcome to this quick introduction to AI and how AI works. But before we dive in, let me briefly introduce myself. I’m Tobias, managing partner of Rapid ai, a German AI consultancy firm that helps B2B businesses use AI to grow their business. And my passion is really to unlock AI to non-technical companies and help them leverage this amazing technology to grow their business. 

So without further ado, let’s dive into the meaning of AI and understanding what AI is, in simple terms. And in simple terms, we can actually say that AI is about creating smart tools that can think on their own, like us. They can mimic human intelligence. And if you think, well, that sounds a bit fuzzy, then actually you are right. AI has been around since the early 1950s as a concept, as a terminology, and from the beginning on it was a very fuzzy term, and it still is. Overall, the definition of AI is to try to create systems, computer systems that are able to mimic human intelligence. And again, we will take a look at what this actually means and what’s going on under the hood. But for now, you can just understand AI as a technology that helps you to process very large amounts of data in a way that humans can’t. It is able to learn things and to see and recognize patterns and to make choices and act very similar to humans. And to do all this, AI, of course, needs to have a lot of data incorporated into it. 

And by the way, you’re probably already using AI in your daily life.

For example, if you’re using voice assistants like Alexa or Siri or Google Assistant or recommendation systems in shopping platforms that you’re interacting with on a daily basis. 

So how does it actually work? How can AI try to mimic human thinking? AI mimics human intelligence through a combination of techniques and technologies that enable machines to perform tasks and make decisions in a way similar to how humans do that. And you might have heard a lot of these technologies already in maybe your own work experience and talking with different vendors in evaluating different tools. And some common themes that all pop up here are terms like machine learning, natural language processing, and computer vision. 

Actually, machine learning is the key technology that underpins all of these technologies and all of these concepts because machine learning is nothing else but a paradigm or a technique that helps computers to learn and recognize patterns from historic data. The whole idea is to not tell a computer system an explicit set of rules of how to behave, but to allow computers to figure out these rules automatically upon seeing and reviewing historical data. 

In short, AI is really changing the world in remarkable ways and making Our whole devices and systems that we use smarter and of course, more useful, especially in the advent of latest advancements in generative ai. And you all heard about these chat systems that are very, very accessible and everyone that is able to say something in plain English or in their own language, they can actually use and interact with these systems. 

What’s the secret source that AI is using? It’s two things. It’s data and learning. And I already touched on both concepts previously, but let’s do a quick deep dive here. Data really is the fuel. At the heart of AI’s abilities is a rich source of data. Every AI system that you interact with, has been trained on a large corpus, on a large historical data set. So it was able to figure out and learn some patterns from this data. And AI systems thrive on really massive amounts. And we are talking about terabytes or petabyte scale data here of learning for these huge AI systems that you can interact with, like for example, ChatGPT and so on. Just like we learn from our own experience, data powers AI. Imagine if you are a human, if you are born as a baby, this baby learns their first steps in the world. They see things, they smell things, they listen to things. This is all data. And their brain is able to pick this pattern, these patterns up from the stuff they learn as they grow and as their brains grow. And very similar AI systems with the right architectures and with the right, with the right processes are able to learn more patterns and more behaviors upon seeing and reviewing new data and then adapting their behavior. 

Here’s some real life examples for you. What can we use AI for and what is it used for in business? AI is used for many different purposes and for many different tools. For example, one famous case obviously is self-driving cars. That’s one aspect of the industry that the automotive industry has been working on for quite some time for a very long time. We’re not there yet, but a lot of cars already have these assistant functions in place where you can, for example, use your car to drive on autopilot or cruise mode on highways or where you can use AI systems in your cars that help you navigate the car better. And all these techniques. These are all first steps into this direction of self-driving and autonomous cars. Social media is also one of the big application areas of AI. Every second there are thousands upon thousands of posts of comments of content being published to social media platforms like X or like LinkedIn or other platforms like that. And you have to somehow figure out what to show the users. If you user logs in, you don’t want to simply get the latest updates because the latest updates would be still 1000, 10,000, a hundred thousand updates per second. So you have to kind of filter that down and give the most relevant content to users. And AI obviously is a good technique to learn patterns and behaviors of what every user likes, what the preferences are, every user, and then show the information which is really most relevant to them. So they are engaged on the platform and revisit the platform again and again.

Voice to text tool is also a primary of AI. I’m talking about this video here in a microphone, but now we can use AI in order to get my voice and transcribe that into text. Maybe turn that text into some other medium as we go from there. And this is also a technology which is used or leveraged in, as I already mentioned, virtual assistants like Siri and Alexa, these voice boards where you can give a voice command but also get a voice message or a voice feedback. 

What’s the future of AI? Where are we heading? There are some exciting possibilities ahead. For example, the first major trend is human machine collaboration. I call that augmentation and augmented workflows. A lot of people think that AI is here to automate tasks and automate human jobs, take jobs away from humans, and we never really know to what extent this will happen. But the only thing that we do know, at least right now, is that AI is helping to enhance human powered workflows. For example, if you are, whether you are writing emails, getting yourself organized during the day, creating slide decks, PowerPoint presentations, if you do any strategic planning, if you do coding or writing, these are all tasks where you can interact with an AI to help you generate either better or more output and thus make you more productive and more effective in your workplace. And especially chat interfaces and voice interfaces, make this machine human machine collaboration so natural and so easy. And this is really a big leverage for, for using AI in so many B2B workflows. 

One area that I’m really excited about is autonomous systems and robotics. So far robotics has always been a little bit decoupled of AI progress. We have the researchers working in the robotics industry and researchers working in the AI industry. These have been two different fields, but right now we are at one point where dissection between them becomes bigger and bigger. Especially with the advent of large language models and with, AI services in general for computer vision and natural language processing and making recommendations, all these things that we heard these are now being applied to robotics, making robots behave more in a human way or make robots being able to detect their environment just by looking at objects and using computer vision algorithms to actually figure out what’s going on in their environment. 

There are a lot of promising examples here and more examples that I’m really excited about is AI in education. For example, think about being able to give every student, every pupil out there a customized tutor that is specifically tailored around the needs of the student. That is something which is possible with AI adaptive tutoring. So the tutor is automatically able to adapt to the student’s progress and ask questions, which are the right level between reinforcing learning that they already have and also challenging them with some more and growing through their learning journey. 

And finally, limitless creativity. Also one area which I’m really excited about. I’m also using AI in my daily workflows. I’m writing a lot of content. I’m writing a newsletter every week, and of course I’m using AI to help me with that to outline things, to research things. I have an AI editor that helps me correct spelling mistakes or find any logical flaws in my argument, in my argument. I think there are lots of opportunities here. I’m specifically focused on writing, but for every other aspect of the creative industry, AI is already playing a huge part. For example, for artists using AI to create new images or AI to create videos, music. 

This whole industry and whole community are developing around that, which is leveraging AI in order to be more creative and overall have better impact with their art and grow as a creator here. We really are exploring a future where AI becomes a collaborative partner in creative endeavors and offers us a unique perspective in expanding artistic horizons. 

I hope this presentation has given you some insights on what AI can do, and also got you excited about AI, maybe the way I am excited about AI and I just recommend, approach AI from a very pragmatic angle, see what it can do for you, what it can do for your business. Get on, get hands on, get hands on, try it out, enjoy along the way as you learn and grow. And I wish you good luck on that and a lot of success using AI for your daily work.