Soluna: Powering Renewable AI, Funding the Risk

Soluna offers potentially valuable exposure to the intersection of renewable power and computing, but investors must assume that management can convert early-stage AI projects into contracted, economically viable assets while creating value faster than it dilutes shareholders.

My investment case for Soluna is based on the possibility that access to power becomes one of the principal constraints on the expansion of AI infrastructure.

I assume that locating computing facilities close to renewable-generation assets could provide Soluna with advantages in energy availability, project development or operating cost. This advantage would only be meaningful if Soluna can also deliver the reliability, connectivity, cooling and power quality required by AI customers.

I assume that management can convert a portion of the company’s development pipeline into financed and contracted AI/HPC projects. The reported pipeline should not be treated as operational capacity, and I do not assume that every proposed megawatt will be built.

I assume projects such as Kati 2 and Dorothy 3 will progress from development into binding tenant agreements. At present, Soluna’s financial results remain primarily dependent on cryptocurrency mining and hosting, while reported HPC revenue is immaterial. The AI thesis therefore remains prospective.

I assume that newly energized facilities will produce sufficient gross profit to cover both site-level costs and the company’s considerable corporate and development expenses. Revenue growth alone would not validate the thesis if operating losses and capital requirements continue to rise at the same rate.

I also assume that management can fund construction through a combination of project debt, partners, customer commitments and operating cash flow, rather than relying predominantly on common-equity issuance. This is particularly important because shareholders have experienced substantial historical dilution.

I do not assume that Soluna is undervalued simply because it trades at a lower price-to-sales multiple than selected peers. I assume a higher valuation could become justified only if the company demonstrates binding AI contracts, credible project financing, improving margins and value creation on a per-share basis.

The thesis would be strengthened by:

• a binding AI/HPC tenant agreement;
• disclosed contract duration and economics;
• non-dilutive or limited-recourse project financing;
• evidence of competitive AI power and infrastructure costs;
• sustained site-level gross profitability;
• and slower growth in the fully diluted share count.

The thesis would be weakened by:

• repeated delays at Kati 2 or Dorothy 3;
• further large equity raises without proportional per-share value creation;
• continued absence of meaningful AI revenue;
• rising corporate costs despite additional energized capacity;
• or evidence that renewable colocation cannot satisfy AI reliability requirements economically.

My view is therefore not that Soluna has already established an AI infrastructure moat. It is that Soluna possesses assets and development opportunities that could become valuable if management proves customer demand, project economics and disciplined financing.