AI: what it is and what it is not (yet)

1. AI: what it is and what it is not (yet)

Hi, thanks for joining me on what AI can do and what not yet. Andrew Ng, AI pioneer, famously said "AI is the new electricity." Imagine that for a moment. Just like electricity revolutionized how we live and work, AI will transform our world in ways we're just beginning to understand. Today, we're going to demystify AI, uncover what it truly is, and perhaps more intriguingly, what it is not yet.

2. History of AI

Let's step back in time to 1956, to the Dartmouth Conference, where the dream of AI began. It was during a workshop that the term "Artificial Intelligence" was first coined, marking the official birth of the field. Over the decades, AI has experienced several cycles of AI Summers, periods of intense activity and funding, and AI Winters, times of reduced funding and interest due to unmet expectations. Maybe you remember IBM's DeepBlue defeating chess champion Garry Kasparov in 1997 or Google's AlphaGo victory over Go champion Lee Sedol in 2016. These weren't just victories in games; they were milestones that showed us the potential of AI. And in this era of big data, the possibilities are expanding exponentially.

3. Most important subsets

So is everything smart an AI? AI knows many different subfields, so let's look at the most important ones: Machine Learning (often abbreviated simply as ML): these are algorithms that enable computers to learn from and make predictions or decisions based on data. A subset of Machine Learning are Neural Networks: inspired by the human brain's structure, crucial for developing deep learning models. Deep learning involves a machine analyzing and processing input data through various methods until it determines an acceptable outcome. This approach is also known as machine self-learning. Another example are Large Language Models (or LLMs), currently very popular with the likes of OpenAI’s ChatGPT, Google Gemini and Anthropic's Claude. LLM’s are advanced neural networks capable of understanding and generating human-like text. Last but not least it is important to mention Robotics: This is a branch of technology that deals with the design, construction, operation, and application of robots.

4. Machine learning

But just as we learn in different ways, AI is trained in different ways: Supervised Learning, where it learns from examples. Learning from labeled data predicts outcomes for new, unseen data. Unsupervised Learning, discovering hidden patterns on its own. The AI finds hidden patterns or intrinsic structures in input data. And Reinforcement Learning, learning from trial and error to achieve a goal. Think about the last time you used a navigation app, or how online platforms recommend movies you might like. That's AI at work. AI can already detect when a student's attention drifts in class or predict when a driver is getting too tired. It's in our courts, hospitals, schools, banks and other offices, making decisions, aiding diagnosis, personalizing education, and reshaping how we work.

5. ANI, AGI or Superintelligence?

In principle we are talking about Artificial Narrow Intelligence (or ANI), AI capable of performing specific tasks. According to some, we will reach Artificial General Intelligence (AGI) in the next few decades, when AI will have understanding and cognitive abilities matching those of a human. However a report commissioned by the U.S. government shows that leading AI labs are working toward AGI and many expect AGI to arrive within the next five years or less. This is heavily debated and even the founding fathers of AI do not agree on this, but some even believe that Superintelligence is possible. The moment that AI will surpass all human intelligence. Emerging technologies like Augmented Reality (AR), Virtual Reality (VR), the Metaverse, the Internet of Things (IoT), Autonomous Vehicles (AV), and Cryptocurrency are closely intertwined with AI, pushing the boundaries of what's possible. But we're not just watching the future unfold; we're part of it.

6. Takeaways

As we reflect on AI's journey from its inception to its current state and beyond, it's clear that AI, much like electricity in the early 20th century, is set to redefine our world. As we navigate through the AI landscape, remember, it's not just about understanding the technology but also about envisioning how we can use it responsibly to better our world. I invite you to question, to imagine, and most importantly, to engage with AI in the most responsible way.

7. Let's practice!

Now, let's practice with some exercises!

Create Your Free Account

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.