What is AI?
AI is not a chatbot trick but a new way of solving problems: learning from data instead of programming rules.
Artificial Intelligence
Artificial intelligence: waiting is a strategy, too.
Just probably not a very good one.
Artificial intelligence is software that learns from examples instead of just working through fixed rules. It is changing how companies develop, decide, communicate and work. We help you make more of it than a ChatGPT account and three workshops. Here you can read what this means for your company and where to start. Strategy. Implementation. Operations.
Dartmouth / Birth of AI
John McCarthy
„John McCarthy (computer scientist) Stanford 2006 (272020300)“ by null0 from Singapore, Singapore, 2006
Source: Wikimedia Commons · Licence: CC BY 2.0 · Image credits
Editing: Cropped (image detail), scaled and compressed; otherwise unchanged. Edited by Today is Life.
In 1955, John McCarthy gave an idea a name: “Artificial Intelligence”. A year later, the Dartmouth conference brought together researchers who were convinced that machines could develop abilities that until then seemed reserved for human thinking. For almost seventy years, much of this stayed research, vision or promise of the future. Today the technology has arrived in everyday work. The decisive question is therefore no longer whether AI will change companies. But who learns early enough to change companies with it.
Why we see it this way
The statement “now or never” is of course laying it on a bit thick. We know that. It’s just that, sadly, it gets a little less thick with every month. When ChatGPT became public at the end of 2022, it was first about texts, then images, then programming. Now we are talking about systems that don’t just deliver an answer but are given tasks, gather the information they need, use other systems and keep working afterwards. While we are still discussing whether an AI writes the better text, the real development has long moved on.
What is unsettling is not even just the technology itself. It is above all the pace. In the past, with a new technology you could quite reasonably say: we’ll watch this for two years first. Then the teething troubles were gone, others had paid for the lessons and you could decide in peace. With AI, two years are currently half an eternity. Much of what we wrote a year ago about what AI can do, we would word differently today. Not because it was wrong, but because it now seems almost banal. And that is exactly what makes “let’s wait and see” so difficult. Wait for what, exactly? The finished model? There won’t be one. The final technology? Neither. The moment when someone says: right, AI is finished now, companies can start? That is very unlikely to come, either.
We still use AI like software
That is where it gets interesting for us, because at the moment many companies still treat AI like classic software. There is an existing process, and somewhere a little box labelled “AI” gets added. The employee now writes their email with ChatGPT, the meeting is summarised automatically and in customer service a bot answers the first three questions. Fine. Often even saves money. But if that is where it ends, we have probably missed a pretty large part of the point. The truly exciting question is whether we would still build this process like this today at all. If an AI can read a customer enquiry, knows the customer, finds their previous cases, gathers the necessary information from different systems, drafts a reply, documents the case and then triggers the next step: why are we still building processes in which people gather the same information from five windows?
For us, that is AI-first. It’s not about as much AI as possible. It’s about putting everything on the table first and rethinking it. And sometimes, by the way, the result is that you don’t need AI at all. But you only find that out if you look honestly anew, instead of just upgrading the old a little.
Waiting is of course a decision, too. The problem is: you don’t see its cost. No tax adviser sends an analysis at year end with the line item “Waited too long: €184,750”. You notice late. One competitor suddenly develops software in months instead of years. The next needs fewer people for its administration. Another answers customer enquiries on Sundays at 22:43, without putting anyone at a desk at 22:43 on a Sunday. At first it doesn’t look dramatic. Until at some point two companies offer the same thing and one of them is faster, cheaper and maybe even better. Then a technology topic quickly turns into a commercial one.
That is why now
The comparison with the internet is a bit overused by now, but it still helps. Of course in 1998 you could say: we’ll watch this internet for now. And of course in 2005 you could still build a website. But by then Amazon, Google and others were long at a very different question. For them it was no longer about whether to “use the internet”. Their business existed only because the internet did. That is exactly the difference that interests us with AI today. Not the question of where one could use AI somewhere. But the much more fundamental one: what would we build differently today if we could start a company again on a blank sheet of paper and AI were simply there?
There is no general answer to that. Luckily. Otherwise we could save ourselves the consulting. But you should start asking the question. We don’t know where AI will be in five years. Anyone who claims to know exactly probably knows more than OpenAI, Anthropic, Google and the rest of the industry combined. But we can see what already works today. And we see how quickly things become possible that we considered science fiction not long ago. That is why we believe now is a pretty good time to take it seriously. Not with the next ChatGPT workshop followed by twelve months of an AI working group, but with your own company. What do we actually do all day today? Why do we do it that way? What of it could a machine take over? What should people keep doing? And what would we not build at all if we could start again today? Those are the questions that interest us. That is why: now.
Understanding and using AI
Our AI-first approach
Those who merely integrate AI work with old processes and a new tool. AI-first means rethinking the process itself.
AI tools at a glance
ChatGPT can do almost anything, Claude writes cleaner code, Gemini knows your Google mailbox. The choice is not a matter of faith.
Workshops & training
Most teams don’t need an AI vision, but three hours of practice with their own data and real tasks.
AI & sustainability
A simple text request now needs roughly 0.2 to 0.4 watt-hours (Google: median 0.24 Wh for Gemini). The problem is the volume, not the single question.
Ready for AI?
Very few companies need a battle of strategy slides. Usually an honest look at three or four concrete processes is enough.
The numbers speak clearly
Up to 50%
productivity gain possible through AI (McKinsey, 2024)
Only 34%
of German SMEs feel prepared for AI (Bitkom, 2024)
300 million
jobs worldwide potentially changed by AI by 2030 (Goldman Sachs, 2023)
Frequently asked questions
What is artificial intelligence?
Artificial intelligence refers to computer systems that solve tasks which normally require human intelligence, for example language understanding, pattern recognition, problem solving and decision-making. Modern AI systems such as ChatGPT or Claude understand and write texts, analyse images and even program.
Does my company really need AI?
Yes. The only open questions are when and how. Those who use AI in a targeted way already have a head start today in efficiency, customer service and innovation. In a few years, AI skills will be as much a matter of course as computer skills are today.
Where is the best place to start?
Start with our workshops. In three to four hours you learn the basics and can use AI tools like ChatGPT or Claude productively straight away. From there we develop an AI-first strategy for your company together.
What does AI consulting cost?
The initial consultation is free. We analyse your situation and show concrete options. Workshop prices start at €149 per person. We calculate individual consulting projects by effort. Before the project begins you receive a transparent quote.
What about data protection with AI?
Data protection is central. We advise exclusively on GDPR-compliant use of AI and show how you can use AI without disclosing sensitive data. The EU AI Act has been in force since August 2024, and the individual obligations apply in stages. As of 15/10/2026, according to the EU Commission’s overview (digital-strategy.ec.europa.eu): bans on prohibited AI practices have applied since 2 February 2025, obligations for general-purpose AI models since 2 August 2025, and the transparency rules under Article 50 from August 2026. For high-risk systems in sensitive areas (Annex III), the deadline was postponed to 2 December 2027, and for AI in regulated products (Annex I) to 2 August 2028. Which obligations apply to you depends on whether you provide or deploy a system and which risk class it falls into – we clarify that case by case.
Ready for AI-first?
No sales pitch with a battle of slides. We look at your company and tell you honestly where AI makes a difference and where it doesn’t. Sometimes the answer is: not yet.