Monaco is a statement before anyone in the room says a word. The cars, the yachts, the hotels, the watches, the access, everything signals something. Wealth, obviously. But also proximity: to capital, to influence, to people who can turn a conversation into a company, a fund or an acquisition.

Spend enough time there during Formula 1 weekend and another thing becomes obvious. In a place built around signals, you become unusually sensitive to the difference between a signal and the thing it is supposed to represent.

That was probably our biggest takeaway from Monaco this year. Not a prediction about markets, and certainly not a claim that a few days of conversations can explain where global capital goes next. This is a more personal account of what we learned, what kept coming up, and what we found ourselves thinking about on the way home.

The conversations shaping tomorrow’s investments often sound surprisingly different from the headlines describing today’s. AI was everywhere, of course. So were robotics, defence, biotech, energy and longevity. But beneath the sectors and the pitches, the same question kept resurfacing:

How do you tell the difference between a business that is executing and a story that has simply become very good at looking like one?

The answer matters because capital has not disappeared. It has become harder to impress.

Investors are asking tougher questions about resilience, execution and whether a business can create lasting value. The interesting part is that these questions are also changing some of the old assumptions about what constitutes risk in the first place.

Four ideas kept coming back to us:

• What we call “safe” is changing
• Narrative has become cheap, while evidence remains expensive
• The best opportunities often appear before they become obvious
• Genuine brave capital means having conviction before consensus, without abandoning discipline.

3. The Biggest Opportunities Rarely Begin With Consensus

You cannot spend much time around investors today without talking about AI. It is the gravitational centre of the conversation and understandably so. But something struck us in Monaco: almost everyone wants exposure to AI. Far fewer people begin by asking what AI needs in order to exist.

That question is often more interesting.

Every technological revolution creates a stack beneath the thing consumers see. Behind AI sit semiconductors, electricity, data centres, cooling, networking, storage, robotics and advanced manufacturing. The application gets the headline. The infrastructure receives the purchase order.

Schneider Electric is a useful example. It does not need to build the next ChatGPT to benefit from AI adoption. As computing requirements grow, data centres need increasingly sophisticated power distribution, cooling and energy-management infrastructure.

There is also an important difference in risk profile. An application-layer AI company may find its product disrupted by the next generation of models surprisingly quickly. Infrastructure providers such as Schneider supply engineering capabilities that the broader ecosystem requires regardless of which individual AI product ends up winning.

If the gold rush gets bigger, you can bet on the miner. Or you can ask who sells something every miner eventually has to buy. The old “picks and shovels” analogy survives because the economics behind it survive. Today's shovels include semiconductors, energy infrastructure, data centres, networking, storage and advanced manufacturing.

Foundation models illustrate the point nicely. Models such as those behind ChatGPT, Claude and Gemini are trained on enormous datasets and computing infrastructure. Building and competing at that layer has become one of the most capital-intensive contests in technology. Training state-of-the-art models requires extraordinary amounts of computing infrastructure and capital. The competition is enormous and the ultimate economics of many application-layer businesses remain uncertain. But whichever model wins, electrons still have to reach the chips, data still has to move, and heat still has to leave the building.

That is often where an interesting investment thesis starts.

• Not with: What is everyone talking about?
• But with: What does the thing everyone is talking about inevitably require?

AI is already changing assumptions about electricity demand and driving additional investment and debate around nuclear power, including small modular reactors, grid capacity, storage and data infrastructure. It is also forcing new approaches to data transfer, computing infrastructure and the physical systems required to run an increasingly digital economy.

These opportunities are not exactly hidden. They are just simply less glamorous. And that distinction matters, because by the time an opportunity feels completely obvious, you are usually no longer early. The challenge is learning to distinguish between uncertainty caused by genuine weakness and uncertainty caused by a market that has not caught up yet.

That is a much harder question than following consensus. It is also where much of the upside lives.

4. Brave Capital Invests Before Consensus

“Brave capital” can sound dangerously close to a euphemism for reckless capital. That is not what I mean by it. To me, brave capital is informed conviction. It is being willing to fund something difficult before the market has completely agreed that it is sensible, but having enough evidence to explain why you are doing it.

There is still plenty of capital willing to take risk, but it has clustered. AI attracts enormous attention. So do parts of defence and robotics. Meanwhile, equally consequential fields still struggle to attract patient money. And patience itself seems to have shortened.

Scientific breakthroughs used to be understood as things that might take decades. Today, even frontier technology is sometimes discussed as though it should fit neatly inside a five-year fund model. Physics does not care about a fund’s exit horizon. Biology is not especially interested in your next board meeting.

So where is brave capital actually needed?

• Can robotics reduce labour shortages without simply creating a new category of social problems?
• Can space-based manufacturing eventually move resource-intensive processes away from Earth?
• Can energy become abundant without relying permanently on fossil fuels?
• Can medicine shift from extending lifespan to extending healthy lifespan?

Those are the kinds of questions I find more interesting than “what is the next AI app?”

Robotics is a good example. Today's humanoid and dexterous robots still look clumsy. That is precisely why it is easy to underestimate them. People see the robot dropping an object, an engineer sees another training example. Millions of human movements are quietly becoming training data, helping machines learn to grasp, manipulate and interact with objects with increasing precision.

The systems improve. Hardware improves. Models improve.

The question is becoming less “will robots ever be dexterous?” and more “how quickly does the capability curve improve from here?” What looks awkward today may simply be what exponential technology looks like before the curve becomes obvious. A concrete example comes from Budapest. Allonic recently raised funding to industrialise dexterous robotic hands, with backing from angels connected to OpenAI and Hugging Face. It is one small example of a much larger pattern.

Yesterday's laboratory curiosity becomes today's awkward prototype. Today's awkward prototype becomes tomorrow's infrastructure.

The investment question is where on that curve you become willing to believe the evidence.

What Monaco Made Clearer to Us

There is an irony in reaching these conclusions in Monaco. Few places on earth are better at presentation.

Monaco itself is a signal. Arriving there says something. Being invited into certain rooms says something else. The boat, the table, the hotel, the accreditation around your neck, everything has informational value. But informational value is not the same as truth. And perhaps that is why the distinction between narrative and reality felt especially sharp there.

You meet founders who can explain a billion-dollar market in five minutes, family offices that have seen hundreds of versions of the same pitch. Investors who know that the best opportunity in the room may belong to the person speaking least. Policymakers discussing industries whose regulatory frameworks have not caught up with the technology.

You start noticing not just what people are excited about, but what they are quietly sceptical of. And the recurring scepticism was not about innovation. It was about proof.

Is the technology real? Does the team execute? Are customers actually using it? Is the moat technical or just linguistic? Does the opportunity survive contact with regulation? What has to become true for this company to work, and how much of that is already true today?

Those are much better questions than asking whether a pitch sounds exciting.

‍Final Thoughts

We left Monaco less convinced by narratives, not more cynical about innovation. There is a difference.

Capital is still looking for opportunity and there is no shortage of money willing to fund the future. What appears to be getting scarcer is credibility. In a market where a persuasive narrative can be manufactured almost instantly, investors have to work harder to establish what is underneath it. That changes the nature of due diligence. Digital Due Diligence is one part of that shift.

It is not a crystal ball. It is simply another way of refusing to take the presentation at face value. Hiring, engineering activity, customer behaviour, technical output, partnerships and product cadence can help build a more complete picture of what a company is actually doing. And that matters because the most interesting opportunities will rarely arrive labelled as obvious.

Some will look risky because they genuinely are, while others will look risky because the market is still using yesterday's definition of safe.

Some stories will sound extraordinary because the underlying company is extraordinary, while others will sound extraordinary because telling extraordinary stories has never been cheaper.

The investor's job is to know the difference, and perhaps that is the real meaning of brave capital:

not capital that believes everything early, but capital that knows what evidence it needs in order to believe something before everyone else does.