• VIX
    Loading…
  • BIST 100
    Loading…
  • UST Yield 10y
    Loading…
  • S&P 500
    Loading…
  • Brent Oil
    Loading…
  • XAU/TRY
    Loading…
  • EUR/TRY
    Loading…
  • USD/TRY
    Loading…
  • XAU/USD
    Loading…
  • EUR/USD
    Loading…

/

Category

/

Dan Loeb just bet $7 million on AI power, but not how you think

Dan Loeb just bet $7 million on AI power, but not how you think

Photo: Sergei Starostin

Dan Loeb's Third Point LLC increased its stake in Hut 8 Corp. to 1.3 million shares last quarter and disclosed a $3.5 million face-value position in Riot Platforms debt, now worth roughly $7 million. Wall Street read the filing as a vote of confidence in two AI infrastructure companies. The reality is more complicated, and the gap between the two is where the real story lives.

The distinction is not a technicality. Third Point owns Hut 8 common stock, meaning it participates directly in the company's gains and losses. Its Riot exposure is debt, specifically notes with a fixed contractual claim. Debt holders get paid before shareholders when a company runs into trouble. They also don't capture the upside if everything goes right. Treating these as two equivalent bets on the same AI infrastructure theme, as much of the coverage did, misreads what Third Point actually filed.

What these companies are actually doing

Hut 8 and Riot both operate in a space that has become suddenly very crowded: converting access to electrical power into leasable computing infrastructure for AI workloads. This is the business of finding land with grid connections, building or repurposing data centers, and signing long-term leases with companies that need enormous amounts of computing capacity.

Hut 8's numbers are large but mostly prospective. As of August 4, the company reported 949 megawatts of contracted IT capacity and roughly $26.6 billion in expected contract value over the life of those agreements. It has secured $7.5 billion in project financing. These are not revenues already earned. They are a roadmap that requires construction to be completed, tenants to move in, and payments to actually arrive. The financing that makes the buildout possible also creates obligations that will take priority over shareholder returns.

Riot's data-center revenue gives a smaller but more concrete picture of what the business looks like once it's operating. In the second quarter, the company reported $23.2 million in data-center revenue. But $18.3 million of that came from one-time tenant fit-out services, not recurring rent. The stable, repeating revenue base is considerably smaller. Riot has delivered the first 25 megawatts of capacity to AMD and announced a 191-megawatt lease with a frontier AI lab, with delivery phased into late 2027 and mid-2028. That pipeline expands the opportunity and the execution risk at the same time.

Who bears the risk

Short sellers have taken meaningful positions against both companies: roughly 12% of Hut 8's tradeable shares and nearly 15% of Riot's were sold short as of mid-August, according to the source material. That doesn't mean the shorts are right. It does mean a significant number of professional investors are betting that current valuations price in more execution success than will actually materialize.

This is exactly the tension that makes Third Point's instrument choice interesting. By holding Riot debt rather than equity, the firm has a claim that sits higher in the repayment order if things go wrong, while Hut 8 equity holders carry the full weight of whether that $26.6 billion in projected contract value survives contact with construction schedules, financing costs, and tenant demand.

The broader pattern here is one of institutional investors finding ways to get exposure to AI infrastructure growth while managing downside differently than a straightforward stock purchase would allow. Debt, convertible notes, and structured instruments let sophisticated firms calibrate risk in ways that a simple reading of their SEC filings will not reveal.

Third Point's filing is not a recommendation. It is a window into how one experienced investor is positioning across the capital structure of two companies with large ambitions and substantial distance still to travel.