Everyone's talking about AI. Few know what to do with it.
Imagine you are in a strategy meeting. Someone says: We need to do something with AI. Everyone nods. No one objects. And then... nothing happens. Or worse: everything happens at once, without a plan, without prioritization, without measurable goals.
We've seen this scenario in almost every company for three years. Not because those in charge are incompetent. But because the AI landscape is changing so rapidly that even technology teams are struggling to keep up.
Over the last ten years, we've recorded 297 AI mentions in our trend corpus. In 2017, there were exactly 6. In 2025, there were 86 – the highest single value in the AI cluster of our dataset. Growth by a factor of 14 is impressive. But a single number doesn't say anything about business value.
The explosion curve: 2022 to 2026 in numbers
Look at the developments of the last five years. Not as an abstract trend line, but as what they are: a signal that has become too loud to ignore.
Each of these numbers represents companies that have either traded – or discussed PowerPoints.
The jump from 2023 to 2024 – from 26 to 64 mentions – is not gradual growth. That is a regime change. And 2026 already has 59 mentions from a single report. The signal is clear: AI is not coming. AI is here.
A consultancy company that introduced an AI copilot for proposals at the start of 2024 has halved its bid preparation time. The win rate has increased by 4 percentage points. Not because the consultants have become better, but because they are faster. Every quarter you wait, the gap grows.
The vocabulary change that most overlook
What lies between the numbers is almost more important than the numbers themselves. Because the way AI is spoken about has fundamentally changed since 2023.
Before 2023
Neural Network
Deep learning
From 2023
LLM
GPT
Coding Agent
Agentic AI
Before 2023, AI was a topic for data science teams. The keywords – Neural Network, Deep learning – were technical, specialised, and a long way from the day-to-day business of most employees.
Since 2023, they have dominated LLM, GPT, Coding Agent und Agentic AI the discourse. The difference? These new forms of artificial intelligence don't just affect the IT department. They change how sales works, how support functions, how code is written. They are no longer lab experiments. They are tools that lie on your employees' desks today – whether you planned for that or not.
Three high-leverage use cases – now
Enough analysis. Let's get down to practice. Our Innovation Radar identifies three AI use cases that can deliver the greatest business value in the next 24 months. No theory. No buzzwords. Instead, use cases with defined problems, measurable KPIs, and built-in controls.
Evidence A
Sales teams spend too much time searching for previous decks, case studies, and delivery artefacts.
Human Approval
Prompt-Injection Testning
Evidence B
L1 support queues are overloaded with repetitive incidents and inconsistent ticket routing.
Escalation Fallback
Audit Logs
Evidence C
Manual remediation of legacy Python and PowerShell assets is slow and inconsistent.
Compulsory Test Gates
Secret Scanning
Please note the evidence classes: A, B, and C. This is not by chance. The Proposal Copilot has the strongest data base, and the Code Modernisation Agent has the weakest. This does not mean that the third use case is unimportant, but that it requires more validation before you scale.
Equally important: What you should leave out
Strategy is not just about deciding what you are going to do. Strategy is also the discipline to say no. Our radar explicitly identifies what you should stop or postpone.
Stop
Build your own foundation model training infrastructure. The costs are disproportionate to the benefits. Use existing models and invest in fine-tuning and orchestration.
Stop
Isolated chatbot pilots without explicit workflow integration and a sponsor with KPI responsibility. A chatbot without process integration is a demo, not a product.
Defer
Multimodal avatar assistants for external workshops. Technically fascinating, but the maturity level doesn't yet justify the investment.
The stop list is inconvenient. But it protects your budget from the most expensive mistakes: projects that neither solve a measurable problem nor have a clear owner.
Regulation is coming – and that's good news
Many see the EU AI Act as a brake. We see it as a catalyst. Because regulation creates something that the AI market has so far lacked: legal certainty. And legal certainty lowers the risk for investment decisions.
Here is the timetable you need to know:
In power
In power
In 4 months
In 5 months
In 17 months
The message is clear: the regulatory clock is ticking. But it's ticking predictably. Those who establish governance structures now will have a competitive advantage – not a burden.
Governance is not a brake. Governance is an accelerator.
This is not an opinion. This is a hypothesis that our data supports.
If governance controls are embedded from the start – evaluation harness, human approval gates, audit logging – then the pilot-to-production conversion rate more than 40 % within twelve months lie. Initial pilot samples suggest this, even though reliable long-term data is still pending. Companies without embedded controls? They remain stuck in the pilot stage, quarter after quarter.
Teams with model routing, human approval gates, and audit logging scale safely to production – while others remain stuck at the pilot stage. This is not a hypothetical scenario. This is the reality we will observe from 2026 onwards.
The reason is simple: without governance, eventually there comes a point where legal, compliance, or the board of directors stops a project. Then you've invested months and are back at square one. With built-in governance, this doesn't happen because the controls run alongside from day one.
The backing vocals: Cloud and Semiconductor
AI does not exist in a vacuum. Two technologies form the foundation upon which any AI initiative stands – and both are showing a clear signal on our radar.
Cloud Computing It has around 20 mentions over six years and has ADOPT status. But not as a standalone innovation topic. Cloud is table stakes – a prerequisite, not a differentiator. If your company is still discussing the cloud question in 2026, you have a different problem.
Semiconductor shows around 11 mentions, spread intermittently over the years, with the status TRIAL. The chip crisis has shown how fragile the AI infrastructure is. Those planning AI workloads must keep an eye on the semiconductor supply chain – even if that is not a topic for day-to-day business.
Together with AI, this results in a cluster of 328 combined mentions. The three technologies form a system – and should be assessed as a system.
What does that mean for your business – specifically?
If you take away one thing from reading this article, it should be this: AI is no longer a technology topic. AI is a business topic. The question isn't whether you use AI. The question is whether you do it with a system or in chaos.
ADOPT does not mean: everything at once. ADOPT means: Start with a clearly defined use case. Build in governance from the outset. Measure success after 90 days. The three use cases above – Proposal Copilot, Service Desk Triage, Legacy Code Modernisation – are proven entry points, depending on whether your greatest pain point lies in sales, operations, or engineering.
And stop launching isolated chatbot pilots without clear workflow integration. Without a measurable goal and a sponsor with KPI responsibility, AI pilots are a waste of budget in good conscience.
The data is clear. The regulation is predictable. The use cases are defined. What's missing is your decision.