Spokesperson: Alexis Cathalifaud, CEO
As demand for synthetic intelligence compute continues to develop, the infrastructure supporting that demand is changing into a strategic consideration in its personal proper. Firms throughout the sector are racing to safe entry to more and more highly effective GPUs, whereas questions round electrical energy, data-center capability, cooling and connectivity have gotten more durable to separate from the compute itself.
Clichmont is taking a special strategy. Reasonably than constructing its mannequin primarily round rented GPU capability, the corporate is targeted on proudly owning and controlling the bodily infrastructure on which successive generations of AI {hardware} can function. On this interview, Clichmont CEO Alexis Cathalifaud discusses why the corporate believes energy and data-center infrastructure may turn into the extra sturdy bottlenecks, the way it approaches web site choice and the challenges of scaling bodily infrastructure, in addition to the function of its $CLAI token throughout the broader ecosystem.
1) Each firm on this class is preventing over GPU entry proper now. Clichmont’s reply is to construct the info facilities as an alternative of renting the chips. Why does possession matter greater than entry?
As a result of GPU entry provides you compute; infrastructure possession provides you management over the economics of compute.
For a corporation like Clichmont, proudly owning or controlling the data-center layer can matter extra strategically than merely securing rented GPUs. Once you hire GPU capability from a hyperscaler or GPU cloud, you inherit another person’s pricing, availability, energy constraints, networking structure, deployment schedule, and margins. When demand spikes, entry can turn into costly or constrained.
Proudly owning the infrastructure modifications the equation. Clichmont can doubtlessly determine which GPUs to deploy, when to improve them, how densely to put in them, how energy and cooling are engineered, and the way the capability is commercialized. The identical facility can even evolve from one GPU technology to the following relatively than tying the enterprise thesis to a selected chip.
There may be one other essential distinction: GPUs depreciate rapidly; power-ready data-center capability is a longer-lived strategic asset. A GPU technology could turn into economically much less aggressive inside a couple of years, whereas land, grid connections, substations, cooling infrastructure, fiber connectivity and permitted megawatts can stay precious throughout a number of generations of accelerators.
That makes the scarce useful resource more and more not simply the GPU itself, however the power to energise hundreds of GPUs at scale. An organization should buy chips and nonetheless have nowhere appropriate to deploy them. Securing 10,000 GPUs is one downside; securing the tens of megawatts of dependable electrical energy, cooling and community infrastructure required to function them is one other.
2) You’re up towards firms which are already public or heading there – CoreWeave, Crusoe, Lambda. What do you suppose their mannequin will get unsuitable, if something?
I don’t suppose CoreWeave, Crusoe or Lambda received the mannequin unsuitable. They proved that AI compute is an enormous market. The place we differ is in what we consider will stay scarce. GPUs change each technology. The sturdy bottleneck is the infrastructure required to run them — energy, land, cooling and connectivity. Clichmont’s thesis is that relatively than competing solely to hire the most recent GPU, we wish to management the infrastructure on which successive generations of GPUs will function. In a market the place everyone seems to be chasing chips, we’d relatively personal the place the place the chips must stay
3) There’s a rising argument that power, not chips, is the precise bottleneck for AI infrastructure. How a lot does that form the place and the way Clichmont builds?
Vitality shapes nearly each infrastructure choice we make. A GPU with out dependable energy is simply costly {hardware} sitting in a rack. We consider the actual competitors over the following decade gained’t merely be for GPUs—it is going to be for megawatts.
So when Clichmont evaluates a web site, we don’t begin by asking the place we will discover the most cost effective constructing. We ask: the place can we safe dependable energy, on the proper economics, with the power to scale? What’s the time-to-power? What’s the grid scenario? What cooling structure does the local weather enable? And may that web site help the following technology of GPUs, not simply those we’re putting in at present?
That’s one cause places with sturdy power fundamentals are strategically fascinating to us. Chips may be shipped all over the world. You possibly can’t ship 100 megawatts. The compute in the end has to go the place the power is.
So I wouldn’t say chips cease being a bottleneck. They continue to be vital. However more and more, proudly owning GPUs isn’t sufficient. The aggressive benefit is having the ability to energy, cool and function them economically at scale. That’s what we’re constructing Clichmont round.
4) Clichmont’s websites vary from a solar-powered facility in Alicante to a brand new construct in Bodo, Norway. What truly decides the place a knowledge heart will get constructed – is it about power, land, local weather, one thing else?
We don’t select a location as a result of one variable appears engaging. We select it as a result of the whole infrastructure equation works.
Energy is the primary filter: what number of megawatts can we safe, at what price, how dependable is that provide, and—critically—how rapidly can it truly be delivered? Then we have a look at cooling, local weather, fiber connectivity, land, allowing, safety and the power to broaden.
Bodø and Alicante are fascinating exactly as a result of they characterize totally different strengths. Northern Norway provides us a local weather that may help environment friendly cooling and a robust power surroundings. Alicante provides us a special power profile and the chance to combine photo voltaic into the infrastructure technique. We don’t consider each Clichmont knowledge heart must look equivalent—the structure ought to reply to the sources of the situation.
And land by itself isn’t notably precious to us. An affordable parcel with no scalable energy or fiber just isn’t a data-center web site. What issues is whether or not we will flip that location into dependable, economically aggressive compute capability.
In the end, we’re not likely searching for land. We’re searching for locations the place power, connectivity, cooling and scalability converge. That’s the place we construct.
5) That is an infrastructure firm with a token hooked up to it. For a reader who’s skeptical of that mixture, what’s the sincere case for why $CLAI exists in any respect?
The skeptical view is totally honest. A token shouldn’t exist simply because an organization operates in AI. If $CLAI had been merely a financing wrapper round our knowledge facilities, I wouldn’t contemplate {that a} compelling cause to create it.
Clichmont is the infrastructure enterprise. It builds and operates compute capability. $CLAI is meant to be a digital financial layer across the broader ecosystem — one thing that may finally help on-chain participation, treasury exercise and group governance in ways in which typical fairness isn’t designed to do.
And we have now to earn the best to make that distinction. The bodily infrastructure has to exist independently of the token, and the token has to display actual utility independently of hypothesis. If we will’t present each, then the skepticism is justified.
So I wouldn’t ask anybody to consider in $CLAI just because Clichmont owns GPUs or builds knowledge facilities. The check is way less complicated: does the token finally do one thing helpful, clear and measurable that couldn’t be achieved as successfully with a traditional database or typical company construction? That’s the usual we ought to be held to.
6) What’s the toughest a part of scaling bodily infrastructure that individuals who’ve solely constructed software program are inclined to underestimate?
The toughest half is that bodily infrastructure doesn’t scale at software program pace. In software program, if demand doubles, you may usually provision extra capability rapidly. In a knowledge heart, each further megawatt has a bodily dependency behind it — grid capability, transformers, switchgear, cooling, fiber, permits, building and in the end {hardware}.
And people dependencies don’t transfer in parallel as neatly as individuals think about. You possibly can have the land and never have the ability. You possibly can have the ability allocation and wait months for electrical gear. You possibly can have the constructing prepared and nonetheless be ready for a grid connection. One lacking element can delay a complete deployment.
The opposite distinction is that errors are costly and troublesome to reverse. Software program may be patched in a single day. You possibly can’t patch a badly designed 50-megawatt electrical system in a single day. You’re making capital choices at present primarily based on what GPUs, energy densities and cooling necessities could appear like a number of years from now.
So the actual ability isn’t merely constructing knowledge facilities. It’s sequencing capital, energy, building and buyer demand in order that they arrive at roughly the identical second. Construct too early and you’ve got costly idle infrastructure. Construct too late and the shopper goes some other place.
That execution self-discipline might be what individuals coming purely from software program underestimate most. In bodily AI infrastructure, pace issues — however timing issues much more.
7) In the event you needed to title the largest danger in betting on a build-it-yourself mannequin as an alternative of a capital-light rental mannequin, what wouldn’t it be?
The most important danger is capital depth mixed with timing. Once you construct infrastructure your self, you’re committing important capital at present towards assumptions about demand, energy economics and expertise a number of years into the longer term.
A rental mannequin provides you flexibility. If the market modifications, you may scale back capability, transfer suppliers or undertake the following technology of {hardware}. Once you personal the infrastructure, you don’t have that luxurious. A substation, cooling system or data-center constructing is a long-duration choice.
For us, the largest hazard due to this fact isn’t merely spending an excessive amount of — it’s constructing the unsuitable capability, within the unsuitable place, on the unsuitable time. In the event you construct forward of demand, capital sits idle. In the event you construct too slowly, you miss the market.
That’s why we don’t view possession as ‘construct every part ourselves.’ The target is to regulate the strategic infrastructure whereas remaining versatile round expertise. The constructing, energy, cooling and connectivity ought to survive a number of generations of GPUs relatively than changing into depending on one {hardware} cycle.
So sure, the capital-light mannequin has an actual benefit: optionality. Our guess is that if we execute accurately, giving up some short-term optionality creates one thing extra precious over the long run — management over capability, energy economics and the bodily infrastructure that AI more and more is dependent upon.
8) Three years from now, the place would you like Clichmont to take a seat relative to the CoreWeaves and Nebiuses of the world?
Three years from now, I don’t anticipate Clichmont to be the largest firm within the class, and that’s not the target. CoreWeave and Nebius have huge scale and entry to capital. Attempting to copy them could be the unsuitable technique for us.
I would like Clichmont to be acknowledged as one of the environment friendly unbiased AI infrastructure operators in Europe — with actual working belongings, secured energy, high-density GPU capability and a monitor file of bringing new compute on-line rapidly.
Our benefit has to come back from being disciplined about the place we construct and what we personal. We wish places the place the power economics make sense, infrastructure designed round successive generations of accelerated computing, and the pliability to serve enterprise AI, HPC and personal compute relatively than merely competing for GPU rental quantity.”
If CoreWeave and Nebius are constructing hyperscale AI clouds, Clichmont can occupy a special place: a targeted proprietor and operator of compute-ready infrastructure in strategically chosen markets.
Conclusion
Clichmont’s technique in the end comes all the way down to a long-term infrastructure guess: that entry to GPUs will stay essential, however the skill to energy, cool, join and function these GPUs effectively at scale will turn into an more and more precious benefit.
That strategy comes with significant trade-offs. Constructing bodily infrastructure requires substantial capital, lengthy planning horizons and cautious coordination between energy, building, {hardware} and demand. Clichmont’s thesis is that accepting these constraints can present higher management over the infrastructure required for successive generations of AI compute. Whether or not that thesis proves out will rely much less on the ambition of the mannequin than on the corporate’s skill to execute it effectively and on the proper time.