Elorian, a visual AI startup founded by former DeepMind researchers, just closed a $55 million Series A at a $300 million valuation. The headline is designed to impress. But to my macro lens, it reads as a liquidity event—not an innovation signal.
I’ve seen this pattern before. In 2017, my Copenhagen hedge fund flagged the ICO bubble by mapping token raises to global M2. Today, the same indicators flash yellow. The Fed’s balance sheet is still in contraction mode, yet venture capital is pouring into AI narratives with the same fever that once fuelled NFTs and algorithmic stablecoins. This isn’t a technology story. It’s a capital allocation story dressed in DeepMind credentials.
Context: The Global Liquidity Map
The $55 million raise represents roughly 18.3% dilution at a $300 million valuation. For context, the median Series A in AI in 2025-2026 is $15-20 million. Elorian’s round is three times the median. The premium is entirely attached to the founder pedigree and the vague promise of “redefining industry standards.” No product. No revenue. No API. No white paper. Just a team and a press release on Crypto Briefing.
Crypto Briefing is the detail that makes me pause. Why would a visual AI company, whose target market is enterprise and robotics, announce funding on a crypto news site? The most charitable explanation: the investors include crypto-native funds that expect a token or a DAO structure down the line. The less charitable: the entire announcement is a paid PR placement designed to attract retail capital from the crypto crowd. Either way, it signals a detachment from traditional tech PR norms. It signals that Elorian’s backers are playing the same game that launched a thousand DeFi protocol tokens into the stratosphere in 2021.
Core: Elorian as a Macro Asset
Let’s stress-test this valuation the same way I stress-tested Aave’s liquidity pools in 2020. Assume a 50-person team with average comp of $300,000 (Copenhagen-level, though DeepMind talent commands more). That’s $15 million per year in salaries. Add compute: training a state-of-the-art visual model requires 10,000 H100 GPU hours per run, at roughly $2 per hour. A single training run costs $20,000. With daily iteration, compute alone hits $7 million per year. Office, legal, marketing—another $3 million. Total annual burn: $25 million. The $55 million round gives a runway of just over two years.
But two years is eternity in AI. Google DeepMind, the same lab that birthed the founders, is spending over $10 billion annually on compute and talent. Elorian is raising a $55 million round while competing against the most capital-rich entities in human history. The only way this ends well is if Elorian’s “innovation” is so radical that it leapfrogs the incumbents—or if it gets acquired by one of them before the cash runs out.
The acquisition scenario is the real exit. And that explains the $300 million valuation. It’s not based on fundamental revenue. It’s based on the option value of being bought by Microsoft, Meta, or Google for $1 billion+ within 36 months. The premium is a call option on future acquisition, priced during a bull market in AI hype.
Code is law, but man is the loophole. In crypto, we learned that human greed always exploits the cracks in the smart contract. In AI venture, the loophole is the narrative. Elorian’s narrative is pristine: ex-DeepMind, visual AI, “industry standard redefinition.” The $55 million is the liquidity flowing into that loophole. The question is whether the fundamental technology can close the gap before the liquidity dries up.
Contrarian: The Decoupling Thesis
The conventional wisdom is that AI investment is a secular trend, divorced from central bank liquidity cycles. I disagree. I published a guide in 2022 titled “Crypto as a Risk-On Asset Class,” demonstrating that Bitcoin has a 0.7+ correlation with global M2 over 6-month rolling windows. The same relationship holds for VC flows into AI. When the Fed raises rates, risk capital retreats. The AI hype cycle began during ZIRP and is now being sustained by residual liquidity. The moment recession fears trigger actual tightening of corporate balance sheets, the first budgets to be cut will be experimental AI projects.
Elorian’s $300 million valuation is built on the assumption that the liquidity spigot stays open. That assumption is not backed by any macro indicator I track. The US Treasury yield curve is still inverted. Bank lending standards are tightening. And the commercial real estate crisis is about to squeeze regional banks that fund a portion of VC.
The decoupling thesis is a mirage. AI and crypto are both leveraged plays on global liquidity. When the tide goes out, both sectors will be left exposed.
Capital is just deferred trust. And trust, in this market, is a first-in, first-out queue. The investors who entered Elorian at $300 million need the next round to value it at $900 million or more to show a paper return. That requires a new narrative—a product launch, a customer win, or an acquisition bid. If none materialises within 18 months, the queue will empty fast.
Takeaway: Cycle Positioning
The smartest move right now is not to chase AI funding announcements. It is to map the correlation between macro liquidity and the next wave of distressed assets. I am short AI-themed SPACs and long volatility on AI ETFs. Elorian itself is not tradeable, but its valuation represents a canary in the liquidity coal mine.
When the macro cliff arrives, the visual AI companies with real products—like those powering autonomous fleets or medical imaging—will survive. The rest will be written off as overpriced talent acquisitions. Elorian’s $55 million is a bet on talent, not technology. Talent is fungible. Code is law, but man is the loophole.
The market can stay irrational longer than your margin account stays solvent. For now, Elorian is a rational bet inside an irrational system. But I am watching the exit doors.
