From hype to value: how to keep GenAI realistic and relevant

Since the launch of tools such as ChatGPT, interest in generative AI (GenAI) has grown enormously. The applications seem endless: from content generation and software development to strategic analysis. The rise of generative AI (GenAI) and large language models (LLMs) has fundamentally changed the data and technology landscape. But where buzzwords thrive, unrealistic expectations also arise. And that is precisely where the risk lies. However impressive the technology may be, without good expectation management, even the best AI approach will lead to disappointment.

What do we see in practice as a data & AI consultancy? And how can you ensure GenAI delivers lasting value rather than remaining a hype?

Expectation	Reality
GenAI replaces work	GenAI speeds up and supports work
GenAI fits everywhere	GenAI works well in some areas but is unsuitable in others
GenAI works on its own	Without people, GenAI has no impact
GenAI achieves business goals	Business goals remain the priority. AI supports them

1. GenAI speeds up work, but does not replace it

GenAI is a powerful assistant, but it does not own the outcome.

It helps structure information, generate first drafts and make knowledge accessible.  Yet we see organisations getting started without a clear use case or evaluation criteria. They run plenty of tests, but implement little. The result? Trust erodes.

Our advice: Start small and be specific. Choose processes where human review makes sense, such as document analysis, content creation or internal Q&A. Measure value at three levels: efficiency, quality and adoption.

2. Strategy also means knowing where to draw the line

A data strategy is not just about spotting opportunities, but also about making choices.   Not every AI application fits your culture, processes or compliance requirements. If you use AI strategically, you must also be prepared to say: “We’re not doing this (yet).”  That takes leadership and technical oversight.

Our advice: Ask these questions:

• What are our guiding principles for AI?
• How do we ensure transparency, ownership and data security?
• Who is involved in assessing and prioritising initiatives?

3. The impact lies within the organisation

LLMs are getting better, but technology alone is not the deciding factor. The value lies in how you use it.

• How good is the input data?
• Are processes designed for people and machines to work together?  
• Is there room for feedback and adjustment?  
• How quickly does your organisation learn?

Technology moves faster, but people and processes determine success. GenAI only delivers lasting value when it is embedded in your data strategy, including governance, change management and talent development.

Our advice: Link GenAI initiatives directly to concrete goals. Bring domain experts, IT and data science together from the start. Without collaboration, it remains an experiment.

Conclusion: Realism is not a brake, but the driving force

GenAI is not a fad; it is here to stay. But getting value from it requires clear choices. You need to be clear about your expectations and embed the technology in your broader strategy. That takes vision, ownership and the courage to make choices. Not everything needs to happen now. But what you do now should set the direction for tomorrow.

Our advice: Start with your business goals. Look at where GenAI adds value, not the other way round.

OK, but how do you do that?

We help organisations make GenAI part of their strategy in a considered, practical way. Not with a blueprint, but with advice that fits your goals, processes and people. Drawing on our hands-on AI experience, we help ensure the technology moves beyond experimentation and contributes to real business value.

Curious where the opportunities lie for your organisation? We’d be happy to explore them with you.