TL;DR:
AI won’t hand you a revolutionary strategy from scratch – my experiments confirm this. For real innovation, arm your teams with model literacy, challenge AI’s outputs with deep human insight, and champion a culture of critical thinking. Use AI to accelerate research, but own the vision.
Everywhere you look, AI promises to revolutionise how we think, work, and learn. For those of us tasked with building future capabilities – whether in learning, product, or strategy – the question is urgent:
Can AI dream bigger than we can, or does it just reflect our existing dreams back at us with more polish?
I recently ran an experiment to test this. I tasked several prominent generative AI platforms with a significant challenge: re-imagine corporate strategy and capability-building from a completely blank slate. The prompt encouraged thinking from first principles, leveraging the latest research, and outlining what we’d build if unconstrained by past methods.
My core question was simple: Could AI, if given a truly blank page, architect something radically new?For now, the answer is a qualified “no.” Instead of a blueprint for a revolution, the outputs largely formed a high-gloss echo chamber of current trends. This experiment was illuminating, mostly in what it revealed about the current limitations of AI for genuine blue-sky strategic thinking.
Unmasking AI’s Illusions
The AI-generated visions weren’t poor. They were often comprehensive, well-written, and touched on many contemporary priorities:
personalised, adaptive experiences
in-the-flow-of-work support
micro-experiences
immersive technologies (VR/AR)
data-driven insights
The issue? None of this was truly blank-slate thinking. It was a sophisticated collation of what many of us are already discussing or implementing.
Take microlearning paths. An AI might present them as a futuristic revelation, yet any major L&D report from the last five years – and countless platforms on the market – already revolve around this concept. The AI identifies a valid tool but sells it as a fresh discovery.
In short, the models dutifully reflected current discourse. They produced a polished version of today’s advanced thinking, not necessarily a truly new tomorrow. This highlights the importance of model literacy: knowing that these tools excel at reorganising what they’ve seen, but still struggle to invent from first principles.
Glimmers of Deeper Insight
The experiment wasn’t without merit. One AI output, prompted to draw on academic research, surfaced genuinely thought-provoking ideas – notably Physiologically-Informed Learning Systems.
Example pilot
A sales-negotiation simulation tracks biometric data (e.g., heart-rate variability). When stress spikes, the system introduces calming interventions; when a learner hits a flow state, it ratchets up difficulty.
Hypothesis: biofeedback-driven training could raise real-world deal-closure rates in Sales teams
Such concepts hint at AI’s potential – when guided by sharp, curious humans.
So, What Is AI’s Role in Crafting Future Visions?
AI excels at amplifying what already exists; humans excel at breaking new ground. Here’s a pragmatic division of labour:
Super-charged research assistantAI can race through literature reviews, market scans, and environmental analyses in minutes rather than days. Your role is to apply judgement, discard the noise, and spot what’s missing.
Draft and ideation partnerNeed a first draft of a strategy deck or an outline of options? The model can supply a solid starting point. Your task is to refine the narrative, challenge assumptions, and tailor it to your context.
Pattern recogniser – not inventorGenerative models are masters of pattern synthesis; they surface and recombine known elements. Your responsibility is to inject first-principles thinking, ask “why”, and re-architect the fundamentals where necessary.
Mastering model literacy means knowing when to let AI run fast – and when to slow down for deeper human inquiry.
The Impact: Human-Led Vision Drives Real Growth
Blindly following AI-generated visions risks high-efficiency, personalised versions of the past – not pathways to a different future. Outcomes:
Derivative strategies
Missed opportunities for competitive advantage
Innovation stagnation
To boost revenue, enhance delivery, or build critical capabilities, we – the professionals and leaders – must remain the chief architects.
Your Next Steps
Pilot with purposeDefine a clear Minimum Viable Pilot. Specify the exact question AI will tackle and how human critique will shape results.
Measure the “newness”Audit AI outputs. What’s genuinely novel? What’s merely re-articulated? Insert targeted human insight to close the gap.
Cultivate model literacyDeeply understand the behaviours and limits of each model you use. A summarisation engine isn’t a ground-breaking inventor.
AI can map the current landscape with remarkable clarity. It can even highlight overlooked paths – if steered well. Plotting the course toward a truly transformed future? That remains, for now, a profoundly human endeavour.
AI can generate answers, but humans define breakthroughs.
What core assumption in your work will you – and your AI – challenge next?
If you want to take a look at the outputs from the GenAI tools you can follow this link.