“The only branch of science that I know with more BS published about it than AI is the search for UFOs.” 😅 — Dr. Lieven Vervecken, Founder and CEO of Diabatix

Challenge accepted. I’ll try to do better! This quote from Dr. Vervecken should be motivation enough to explain why I am on a quest to understand what’s going on in the Artificial Intelligence sector in the EV and battery space.

Over the past few months, I’ve been researching and talking to various AI experts and companies with AI products across the battery value chain. It has been truly exciting to see how much is already happening in this space.

Most of you are probably using ChatGPT or other AI tools, but do we really understand what AI is and how it is being applied across industries? That’s why I started this AI in E-Mobility podcast series: to educate and learn from those who have been deeply involved in this space for a long time. Together, let’s separate fact from fiction.

Why AI? Battery is still a black box

I’ve worked in the battery industry for almost 10 years. With a background as a PhD in Technical Physics I was used to focusing on the atomic and molecular scale. I’ve been involved with electrochemical modeling of batteries at the sub-cell level and have applied data-driven approaches to predict battery lifetime in EV fleets.  

What I’ve learned over the years is that the battery is still a black box—and that’s what initially excited me about working in this field. There is still so much we don’t truly understand about batteries. It’s a complex electrochemical system that evolves over time, depending on how it’s used. So many processes run in parallel, influenced by external factors like temperature, mechanical pressure, and more. 

This raises the question: Can AI help us better understand batteries and the entire EV ecosystem? As humans, we’re great at setting up experiments and trying to understand the physics behind things, but of course, we tend to design experiments with specific outcomes in mind. Is there a different or better way to approach this? 

I recently came across a study that demonstrates how machine learning and data-driven approaches can revolutionize battery formation, one of the most critical processes in battery manufacturing that directly influences performance. The study explored how AI could propose experimental designs that might not have been considered by humans. Traditionally, the industry believed that low charge currents were optimal for forming the Solid Electrolyte Interface (SEI) during cell formation—a key factor for ensuring battery longevity. However, this AI-driven study suggested that using high currents could lead to batteries with up to 70% longer life, challenging long-standing assumptions in the industry. 

This is just one example. The potential of AI to unlock further insights across the entire battery value chain is enormous. What about the rest of the battery’s lifecycle and the broader EV ecosystem? 

The Basics of Artificial Intelligence

As a foundation, we first need to understand what artificial intelligence actually is—and what it is not. Let me share some definitions from experts featured in the series: 

• “AI is the science of making machines smart. It’s about building things that give intelligence to machines and software, enabling them to think, reason, and create.” —Paul Roetzer, Marketing AI Institute 

• “AI is a transformative tool. It allows us to become smarter, faster, and more efficient. It helps summarize vast amounts of information and allows machines to think, create, and come up with solutions faster than humans ever could.” —Fabrizio Martini, Electra Vehicles 

• “AI for me is a kind of magic tool, which gives you super power. But at the same time, it has no civil liability. It has this extreme good student, which is not even 100 % perfect. And I’m always saying that we need to keep our critical sense and objectivity.” Virginie Turc, Hexagon Manufacturing Intelligence 

• “AI is a smart system that makes decisions like a human, sometimes faster, sometimes better. At some point, AI designs things that humans cannot even create.” —Dr. Lieven Vervecken, Diabatix 

• “I see AI as an incredibly useful sidekick that helps draw meaningful insights from vast amounts of data quickly. AI finds signals in the data that you may not even know are there, and it helps make sense of them.” —Hilary Taylor, WattsUp 

• “People will say like, AI is machine intelligence, but for me it’s building systems that take over tasks that for us humans have little to no utility.” Paul-Miki Akpablie, Sankofa 

• “AI maps correlations between inputs and outputs. It’s about showing the computer many examples of what you want it to do and letting it learn from those examples. AI can now do things we can’t, because it identifies correlations beyond human capacity.” —Dr. Jason Koeller, Chemix 

As you can see, everyone has their own definitions, but what helped me grasp the concept of AI better was recognizing it as a hierarchy. This hierarchy starts with algorithms as the foundation, progresses into artificial intelligence (AI), and then branches into machine learning—a subset of AI—culminating in deep learning as the most advanced form of machine learning. 

• Algorithms: These are step-by-step instructions that a computer follows to complete a task. Think of an algorithm as a recipe for how to solve a specific problem. 

• Artificial Intelligence (AI): AI refers to machines or software designed to mimic human intelligence. For example, voice assistants like Siri or Alexa that can understand and respond to your requests. 

• Machine Learning (ML): This is a subset of AI where machines learn from data to improve their performance over time. A practical example would be recommendation systems on Netflix, which get better at suggesting shows the more you watch. 

• Deep Learning: A further subset of machine learning, deep learning uses neural networks (layers of algorithms) to process complex data. A good example of deep learning is image recognition, such as facial recognition on smartphones, where the AI learns to recognize faces based on patterns. 

History of AI

Historically, while algorithms have been talked about since ancient times (before Christ), the modern definition of an algorithm came in 1936 with the invention of the Turing machine. This machine introduced the idea of computation and laid the groundwork for what we know today as computer science.

Surprisingly, AI was formally introduced as an academic discipline over 70 years ago, in the 1950s. Back then, AI systems were mostly rule-based, relying on logical “if-then” statements to simulate human decision-making.

In the 1980s, we witnessed the rise of machine learning and the rebirth of neural networks. Machine learning shifted the focus from rule-based logic to enabling computers to learn from data. Neural networks, inspired by the structure of the human brain, allowed machines to recognize patterns and improve over time with experience. This was a critical shift in AI development, moving toward more data-driven systems.

Today, the most prominent form of AI is data-driven, meaning AI learns from vast amounts of data to make predictions and decisions. However, the real tipping point came in 2017 with the introduction of large language models and transformer networks. These architectures, such as the ones behind ChatGPT, revolutionized AI by giving machines the ability to understand and generate human-like text. This breakthrough made AI not only more powerful but also more accessible to non-technical users. Suddenly, anyone could interact with AI systems to generate text, code, images, and more—marking a major leap in the democratization of AI.

Different Types of Artificial Intelligence

As I throw around terms like machine learning, deep learning, and neural networks, I realize that many people, including myself at times, tend to use them interchangeably. But the truth is, these are distinct concepts, each fitting into a broader hierarchy of AI. To make things clearer, I wanted to create a structured visualization that lays out the different types of AI and where they fit within the overall AI landscape.

One important takeaway is that not all AI is data-driven, even though the most popular and widely known examples today—like machine learning and deep learning—are. Other types of AI include things like swarm intelligence, genetic algorithms, and rule-based systems.

And then deep learning is not the only form of machine learning. There are many more approaches:

• Supervised learning is when AI is trained on labeled data. For example, showing an AI thousands of photos of cats and dogs, each labeled as either “cat” or “dog,” so it can learn to distinguish between them on its own.
• Unsupervised learning is when AI is given data without any labels and has to figure out patterns by itself. For instance, an AI might be given a dataset of customer purchase behaviors and group customers into clusters without being told what the groups are.
• Reinforcement learning works through a reward system. The AI learns by taking actions in an environment and receiving feedback, like a game where it gets points for making the right moves. For example, teaching a robot to walk by rewarding it for moving in the right direction and correcting when it falls.
• Deep learning is a more advanced type of machine learning that uses neural networks with many layers. It’s particularly good at recognizing patterns in large datasets, such as identifying objects in images or translating languages. For example, deep learning powers image recognition systems like facial recognition on smartphones.

How is AI Applied and What are the Benefits?

Now that we’ve covered the basics, let’s dive into how AI is actually being applied. To make things clearer, we can break down AI applications into three categories: personal use, organizational use, and industry-specific applications in the E-mobility sector. I’m borrowing this concept from the Aurelia Academy Leadership training I attended earlier this year.

1. AI for Individuals

For many people just starting out with AI tools like ChatGPT, they may try a few things but often struggle to fully grasp the value it brings. We know there’s value, but how do you measure it? Now, ask yourself this: could you stop using AI today? What impact would that have on your daily life?

To give you some real-world examples, here are insights from our interview series:

• “I use ChatGPT all the time—it’s my number one go-to. I also use Perplexity, which has replaced about 80-90% of what I used Google for. Personally, I think of ChatGPT as a strategy assistant, though I still prefer writing myself.” —Paul Roetzer, Marketing AI Institute 
• “I use AI tools like Claude from Anthropic to learn, write code, and summarize long documents. These systems have become incredibly useful.” —Dr. Jason Koeller, Chemix 
• “AI simplifies personal tasks for me, like planning, organizing, and scheduling.” —Virginie Turc, Hexagon Manufacturing Intelligence 
• “AI helps me streamline tasks like writing, research, and data management. It’s an invaluable tool for staying on top of both personal and professional responsibilities.” —Fabrizio Martini, Electra Vehicles 
• “I use AI to prepare texts, presentations, and marketing materials. It helps me save time by summarizing and organizing information.” —Dr. Rafael Fietzek, Compredict
• “At home, I use AI for simple things like Amazon Alexa and Roomba. AI is quietly ingrained in our daily lives, even when we don’t notice it.” —Hilary Taylor, WattsUp 

2. AI for Organizations

When it comes to organizations, AI’s potential is vast. It’s not just about individual productivity anymore—AI can streamline complex processes, enhance decision-making, and create efficiencies across entire operations. In many companies, AI is embedded into their core strategies, helping to automate tasks, improve workflow, and drive innovation. For more in-depth insights on how AI is transforming industries, I highly recommend exploring the work of Dr. Angelika Berger-Sodian, whose keynote I attended during a Battery Manufacturing Conference in Barcelona.

Here are some examples of how AI is transforming organizations from our interview series:

• “In my companies, we use AI for strategic planning and ideation. Tools like Descript automate podcast production, saving us 15-20 hours per week.” —Paul Roetzer, Marketing AI Institute 
• “At Chemix, AI is deeply integrated into everything—from coding and document summarization to the battery development process itself.” —Dr. Jason Koeller, Chemix 
• “AI supports operations in many parts of the company, from optimizing business processes to assisting development teams by automating repetitive tasks.” —Fabrizio Martini, Electra Vehicles 
• “At ComPredict, we’re careful about using generative AI in core areas, but we do use tools like ChatGPT for less sensitive tasks like preparing presentations and summarizing information.” —Dr. Rafael Fietzek, Compredict

3. AI in the E-Mobility Industry

Now, let’s explore how AI is transforming the E-mobility and battery industries. A study by McKinsey from 2018 highlights just how significant AI’s impact is expected to be in the automotive sector by 2025. The report shows that AI applications could generate USD 215 billion in value across the automotive OEMs’ value chain. The areas of highest potential include manufacturing, research and development, and supply chain management, with AI-driven solutions improving everything from production quality control to supply forecasting. I think a lot of this is true and has become reality already.

As Paul Roetzer points out in our first episode:
“Google announced something called AlphaChip. They’re using AI to design their chips, and they believe that in the near future, AI will handle the entire chip design process. Now, if you think about applying that same logic to battery design or car development, we’re moving toward a future where AI could function as an engineer—finding entirely new ways to design batteries, or collaborating with engineers to unlock innovations we haven’t yet imagined. This might already be happening at the most innovative automotive companies, though we may not be hearing about it just yet.”

That’s exactly what I am covering in this podcast series. Companies like Chemix, Monolith Hexagon Manufacturing Intelligence, Diabatix, Electra Vehicles, Texas Instruments, Compredict, WattsUp, and Sankofa are leveraging AI to push the boundaries of E-mobility and battery innovation.

Stay tuned as these interviews are rolled out on the Electrified Veronika Channels.

The Future of Electric Vehicles and Batteries with AI Innovation

As AI rapidly advances, it’s clear that its role in the EV and battery sectors will only grow.

The chart below illustrates how quickly AI is surpassing human capabilities in tasks like speech and image recognition, reading comprehension, and even code generation.

The speed at which AI is mastering these tasks is accelerating, meaning we’re not talking about innovations in the distant future—major breakthroughs are likely just months away.

To stay ahead of these rapid developments, I encourage you to watch the entire AI in E-Mobility series. It covers how AI is already transforming key areas of the industry and what we can expect in the near future. Hope you enjoy!