A few weeks ago, I led a strategic communications session for a cohort of early-stage start-ups, many of them in the techbio space, as part of the KQ Labs programme. I was followed by a brilliant talk from Jean-Baptiste Cazier Head of Bioinformatics & Biostatistics, and Data Science Strategy at The Francis Crick Institute, entitled Beyond the AI Hype, which raised a deceptively simple challenge: can you explain your business model without mentioning AI? And do you really mean AI or do you really mean machine learning, or are you using AI to make predictions?
It’s harder than it sounds, and if you’re looking for financing, it matters more than you think.
We’ve been here before
There has been an exponential rise in AI-driven companies across all sectors, but particularly in healthcare & life sciences. If it leads to faster drug discovery, better diagnostics for earlier disease prediction and better medicines, that’s genuinely exciting. However it has also led to the belief that simply attaching “AI” to a company name can inflate perceived value by ten times or more, and for those of us who lived through the internet boom and bust of the 1990s, this feels uncomfortably familiar.
Back then, company valuations soared on the promise of the internet, until the bust exposed a rather obvious truth: the internet was a powerful new channel, but it couldn’t turn a weak business into a strong one. The same logic applies here. AI, and all its permutations, is a tool, a formidable one, but a tool nonetheless.
The investor in the room
I’ve spent 25 years helping some of the brightest people in the life sciences sector ‘speak human’ when talking to investors and potential partners. The aim is never to dumb down the science. It’s the realisation that many of the people sitting across the table studied business or economics, not biochemistry. The people they are meeting are smart, commercially sharp, and almost always short on time. Within the first minute of a meeting, they are forming a view: is this business model credible, could it make money and by when?
The first meeting is the foot in the door moment, so keep it simple. What problem are you solving? How big is the market? What’s your edge? What does a good outcome look like? If you can land those answers clearly and quickly, the deeper conversation about your science and technology can follow in a second meeting, if there is one.
The AI trap
The same principle applies to the AI question specifically. Start talking about machine learning models, neural networks or agentic systems, and you may well lose your audience in the first minute; it’s the same with me the moment anyone mentions pensions. However, the bigger risk is if an investor does push back and asks what kind of AI you’re using, and your CEO can’t answer that clearly, you’ve lost their confidence. I’ve watched this happen, with the CTO sitting alongside looking agonised.
AI is not one thing, it is a multitude of tools, from machine learning, deep learning, to natural language processing, computer vision, generative models and so on. Knowing which your company actually uses and being able to explain why it matters in plain English, is essential.
The exercise worth doing
So, before your next pitch, try this: describe your company without using the word AI. If you can explain the problem, the solution, the market and your competitive edge in plain language, then the technology becomes something you introduce naturally, rather than something you lean on.
And if you’re still tempted to lead with AI, here’s a thought: Isaac Asimov’s ‘I Robot’ published in 1950 predates the coining of the term Artificial Intelligence by a few years. The concept isn’t new; what is new is your science, your insight, and the specific problem you’ve chosen to go after. You should be leading with this to win over hearts and minds.
Finally, if you’re still finding it hard to articulate why your business is the next best thing, with or without AI, come and talk to us; we might be able to help provide clarity from complexity and help you open more doors.
