Schön, dass Sie wieder mit dabei sind. Bei einer weiteren Folge von TJS Technology Tuesday. In dieser Episode einmal einen Vortrags Ausschnitt einer Kino, die ich vor kurzem gehalten habe zum Thema Künstliche Intelligenz. Ich wünsche Ihnen viel Spaß dabei.
Beginn der Keynote:
Thank you very much and a very warm welcome to my birth city, Munich. As much as I love Munich, I also love AI. But AI is bullshit. 95% of all AI implementation projects fail. This is what current studies from McKinsey and the MIT are clearly stating. And what are the reasons for that? One of the main reason is timing. Because a lot of managers are asking the right questions, but on the wrong end. Like the question, TJ, what are the latest new shiny AI tools. What you see here behind me is an overview of the major AI tools two and a half years ago. Even at that stage, you wouldn’t have been able to even read all of those logos. Now it has grown exponentially. Looking into those millions of tools and gadgets, how is technology usually coming into organizations?
Planned?
Strategically? No. Via the golf course. I always call it the Blackberry effect. Because at one stage, it was quite uncool to show up at the golf course with the Blackberry. So managers went back to their IT Department and said, I need this new iPhone, including those who are always asking their assistance to print out their emails. Find the mistake. And then I see some managers that have more gadgets than they can even carry. And looking into some board meetings, they sometimes remember me more, sometimes to a playground like this. For those of you who have seen my slides before, you might wonder to say, Okay, these are different slides than TJ provided before. Yes, of course. Because I joined you Wednesday in the evening, and I’m very grateful for all those insights I was able to gain through a lot of talks with you during the day and the evening. Because I’m not home really home in your sector, in your industry. I will say, Give me the best tool of the world and send me on the roof. You can just pray that it doesn’t rain. Or when you ask me, What’s a beta?
Of course, a beta is an early stage of software. When you ask me, What’s a zebra? A zebra, of course, is a striped horse. But I’ve learned that probably some terms have a different meaning in your industry. We already talked about the quality of output of artificial intelligence. I always say it doesn’t start at the end, it starts at the beginning. You probably all have heard the saying, A fool with a tool is still a fool. I always extend that in times of AI because I say, A fool with artificial intelligence makes the disaster faster. Artificial intelligence is nothing new. Actually, when did it first come to me? Exactly, and I looked it up, the first book I read on artificial intelligence was published exactly 40 years ago. You see, I’m already a bit longer young. So that was my first book. Coming with my first computer, that was when I was 14. And a nice incident, this photo, I took this in the former Nixtop Computer Headquarter, which is now an IT museum. That’s where I started my working career, 1988. Since then, I’m also professionally asking me the question, how can you use technology to be more productive or to sell more.
From 2000 to 2010, I was responsible as a managing director of a German futures company and also was responsible for 13 European distributors. What my mission is, my mission is to bridge the gap between technology, which is one of my loves, and management, which is the other love, to bring them together because I see technology and management is far too often not interconnected. Since 2010, I help other companies like the Coca-Cola Company, for example, to be more productive with digital technology. Before I continue, I’m very interested in how AI-ready are you. I already started where I’m seeing me from no clue to number 5 being a fully integrated AI company. You are here and your company is already driven by AI to 100%. I’m not there yet. But like yesterday, for example, I sat together with a group of Dutch colleagues and they said, Okay, we’re a group of seven. Three are here, one is on vacation. So literally half of the company was there. Okay, with AI, no problem. Great to see where you are already standing. I see we already have a number 5. Wow, that is great. So somebody who is already an integrated AI company.
I’m very much looking forward to see all of you and especially the edges of the scene, but also all the other ones in my workshop after this keynote, because we will have a break and afterwards I will host a workshop. During that workshop, what you can do is always also ask questions. You should now see on your smartphone the possibility to enter questions, like for example, the one which I put in here, what’s the difference between ChatGPT and Microsoft Copilot? Use the chance during During my talk. It will be open during my whole talk when you have any questions, just key them in. Feel free to also rate them up. When you say this is a very important question, because then we see the highest rated questions first, and then we will go through them in my workshop, and of course, afterwards, because I’m here for the whole day for you. Looking into how many organizations are looking into AI, it’s like this picture where we We have an environment which is still quite conventional, paper-based and with manual fire. Then AI is seen as the magic miracle in that. Just two weeks ago, I was talking to a friend of mine who is the managing or is one of the board members of Goodmills, one of the leading European groups of mills.
So Diamond is one of the brands probably some of you know. They had a company in their house in the in the headquarters that is Palantir. I assume some of you have heard about Palantier. Not only as a provider of military-grade AI, but also as a provider of solutions for large corporations. Their idea was to say, Okay, we want to have AI to streamline our quality management. They said altogether, and at one stage, Palantir got very silent. They said, We are very sorry, but we have to come back in approximately one and a half years because this is what we see in your data. The best artificial intelligence can’t work without any structured data. I even add that in terms of garbage in, garbage out. I see everybody needs, We need AI, we need AI. I say, Okay, probably it’s a good idea to start with common sense and to start with structured systems. When you’re doing analysis, you probably don’t want to have a different analysis result every time. So probably it’s a good idea. Even yesterday in the AI workshop, I learned to say, Okay, don’t let AI do the calculations. Use AI to enrich the deterministic systems with benchmarkings.
What do we expect on about and with AI? We expect it to be like the Terminator. We’re not 100% sure whether it’s a good thing or a bad thing, but we all say it’s perfect, it’s the perfect machine, and it knows everything. Then we’re talking to Siri, and this is like this one to say, Okay, probably who has installed this on my cell phone. Some of you might say, Okay, this is exactly the quality. It wouldn’t have been a need to even mark it as AI-generated. We see that. Ai-generated, we see that it’s AI generated because it’s not perfect. This was a picture which I built one and a half years ago. This one was two months ago. The only prompt I gave the system was, Generate a picture of Steve Jobs with a quote, If you want to make people happy, sell ice cream. What do you see here? I see a lot of people that did some AI one and a half years ago and it gave you results like this. Then I said, Okay, I have been at AI, so it doesn’t make any sense. This is what we were two months ago.
Ai I will never be as bad as today. Looking into the development pace, we’re not talking about years, we’re not talking about months, we’re talking about weeks where we have exponential increasing systems and power. Also, when you look into robotics, for example, so you probably have all seen those pingpong videos where we try to beat the system, we try to be smarter than the robots, and you see, as hard as he drives, the robots are just the cooler ones. But this is AI generated. This is the original video. 100% of these videos which you see on Instagram where robots that are defeating human beings on ping-pong, 100% AI generated. We are in terms of robotic. We are not there yet. But when you see what’s already capable in terms of video generation, we are far ahead of what most of you already imagine. Just to give you an idea where we are at robotics at the moment, this is approximately the pace of the robotics. People say, Did you put some slow motion effect on it? No, this is the realistic pace you have of systems which are in production, like here in the prototype in the BMW plants in the US.
But again, there is a mistake in the system because what we try to do is we say, Okay, we have learned that with the beginning of the Ford Model T, that this is how we produce with humans, and now we try to just replace the human one by one by robots. People who have understood this system, they say, No, our robots are looking like that. Yes, Tesla is building robots, but what they understood in their production is to say, Why shall we take all the limitations that we humans have so when we can build robots that are much stronger, much powerful than humans. By the way, a former German company, Cuker, now owned by Chinese. Also looking in terms of when we’re looking into artificial intelligence, today, it’s a good idea to always look into West direction, USA. When we’re looking into the future of AI, we probably should be looking more eastwards. Who has heard about Deep Seek? I see some hands here. That was the Chinese Artificial Intelligence System. How did it arise? They were the first ones with the reasoning mode. Reason mode means that the artificial intelligence is not just thinking shortly but longer with some iterations.
They did that out of a bottleneck because due to the export restrictions, they had a limited availability on those chipsets. What did they do to say, We don’t have the hardware here. What do we have to do? We just have to be smarter and we have to think harder. Then they integrated the reasoning concept. The US was smart enough to say, Okay, now we have the chips and we can copy the reasoning thing. But the Chinese were first to go. While we in Germany are blocking artificial intelligence in our school because we say it’s a bad thing, kids have to think themselves, AI is starting in class number one in school since years. So the future will be coming from the east.
Ende der Keynote
Fazit
Diesen Vortragsmitglied mitnehmen. Sollten Sie für Ihre nächste Veranstaltung einen informierenden, einen inspirierenden, einen bewegenden Redner suchen, der Ihre Mannschaft, der Ihre Vertriebspartner, Ihre Servicepartner begeistert, inspiriert und vor allen Dingen in die Umsetzung bringt, dann freue ich mich sehr. Wenn Sie mich ansprechen, dann spreche ich gerne auch als Keynote Speaker, als Workshop-Verantwortlicher oder in anderen Formaten für und mit Ihren Kunden. Mit Ihren Kunden spreche ich natürlich dann auch nicht nur diese eine Stunde, wo ich einen Vortrag halte, sondern ich bin bei den Vereinstaunen den ganzen Tag mit dabei, sodass ihre Teilnehmenden hinterher sagen: „Wow, das war aber ein toller Service. Genau das ist meine Idee, seit ich 1988 bei Nixdorff begonnen habe, das Thema Digitalisierung, Vertrieb und Produktivität zusammenzubekommen, aber das Ganze auch mit Spaß, mit Motivation, sodass neue Technik keine Angst erzeugt, sondern dass die Lust auf Digitalisierung, Lust auf KI haben. Ich freue mich sehr, wenn ich auch Ihre nächste Veranstaltung dementsprechend bereichern kann.
Ihr Thorsten Jekel.
