IT'S AN AI BIG BET: VALUATIONS ARE AT A SPLIT STAGE

By Nancy, PANews

 

ENCRYPTION ASSETS CONTINUE TO WEAKEN, AND ENCRYPTION MINING IS FACING INCREASING SURVIVAL PRESSURES. IN AN ATTEMPT TO FIND A NEW GROWTH CURVE, MORE AND MORE MINING COMPANIES ARE ACCELERATING INTO THE AI TRACK, A NARRATIVE OF TRANSITION RAPIDLY GAINING POPULARITY IN THE CAPITAL MARKETS, WITH A NUMBER OF MINING COMPANIES EXPERIENCING SHARP INCREASES IN STOCK PRICES AND EVEN UPDATING HISTORICAL HIGHS。

HOWEVER, WHILE AI ' S OPERATIONS INJECT NEW GROWTH IMAGINATIONS INTO MINING COMPANIES, THE HUGE CAPITAL EXPENDITURES BEHIND THEM, THE SUSTAINED INVESTMENT OF FUNDS AND THE LONG RETURN CYCLES ARE PUSHING THEM TOWARDS ANOTHER FINANCIAL DRAIN. THIS HIGH BET OF TRANSITION TO AI IS TESTING THE FINANCIAL CAPACITY AND IMPLEMENTATION CAPACITY OF MINING COMPANIES IN THE CURRENT PERIOD OF CONTINUED PRESSURE ON THE PROFITABILITY OF MINING OPERATIONS。

The stock price was a big win for bitcoin, and the mineral valuation was at a split stage

MINING COMPANIES, WHICH ARE TRANSITIONING INTO THE DYNAMIC OWNERS OF THE AI ERA。

AS BITCOIN'S PROFIT-MAKING SPACE SHRINKS, SOME MINING COMPANIES EVEN FALL SHORT, WHILE THE AI OUTBREAK DRIVES A SHARP INCREASE IN GLOBAL DEMAND FOR DATA CENTRES, ELECTRICITY RESOURCES AND GPU COMPUTING POWER. AN INCREASING NUMBER OF MINING COMPANIES HAVE BEGUN TO ACCELERATE THEIR TRANSITION TO THE AI INFRASTRUCTURE SECTOR IN SEARCH OF A NEW GROWTH CURVE。

This transformation has a natural advantage for mining companies. In order to meet the demand for large-scale mining, mining companies have long acquired such critical assets as abundant electricity resources, land reserves, power transformer capacity and mature heat and cooling systems. The data centre operator, who started building from scratch, is not the sameBY UPGRADING EXISTING FACILITIES, MINING COMPANIES WILL BE ABLE TO QUICKLY ENTER THE AI INFRASTRUCTURE MARKET IN ORDER TO MEET THE DEMANDS OF AI IN A LOWER COST AND SHORTER CYCLE。

SINCE LAST YEAR, THE PACE OF THE TRANSITION OF MINING COMPANIES TO AI HAS INCREASED SIGNIFICANTLY. SOME OF THE MINERS DECISIVELY DILUTED OR EVEN WITHDREW FROM TRADITIONAL MINING OPERATIONS, MOVING TO FULL-SCALE AI COMPUTING AND DATA CENTRE OPERATIONS; OTHERS RETAINED SOME OF THEIR MINING OPERATIONS, BUT GRADUALLY SHIFTED THE FOCUS OF RESOURCE ALLOCATION AND CAPITAL EXPENDITURE TO AI. TODAY, MANY MINING COMPANIES HAVE GROWN INTO IMPORTANT PLAYERS IN AI INFRASTRUCTURE DEVELOPMENT。

In terms of transition time, CoreWeave, Applied Digital and Bitdeer started setting up AI Calculus and Data Centre operations as early as 2022-2023 and were among the industry's transitioners, while Iris Energy, Terawulf, Hut 8, Riot Platforms, Bitfarms and others began full-scale coding AI infrastructure construction in 2025, just as the AI industry entered a rapid expansion cycle。

In terms of stock price performance, the market has given high recognition to the AI transformation narrative of mining companies. Eleven mining companies increased by an average of 75.97 per cent at the beginning of the year to date, significantly winning bitcoin over the same period, and mostly after transition prices even reached new highs. Of these, Bitfarms (129.62 per cent), Hut 8 (131.87 per cent), Terawulf (118.68 per cent) and Riot Platforms (93.71 per cent) were particularly prominent and were beneficiaries of the current AI infrastructure revaluation。

In terms of market value, there has been a marked fragmentation of the mining industry, with the market value of CoreWeave reaching $62,855 million as a representative of the success of the transition, well above other mining companies and becoming the new valuation pole for the industry; the companies Iris Energy, Terawulf, Hut 8, Appleted Digital and Riot Platforms have formed a market value ladder of $10 billion to $20 billion; and enterprises such as MARA Holdings, Core Scientific, Bitdeer, CleanSpark and Bitfarms remain below $5 billion. This divisionIN ADDITION TO ITS PRE-EMPTIVE ADVANTAGES, THE MARKET HAS BEGUN TO PRICE DIFFERENTIALLY THE AI STRATEGIC IMPLEMENTATION CAPACITY OF INDIVIDUAL MINING COMPANIES, CLIENT RESOURCES AND DATA CENTRE LANDING PROGRESS。

IN BASIC TERMS, HOWEVER, MOST MINING COMPANIES ARE STILL AT THE STAGE OF REINVESTMENT IN THE AI TRANSITION. ALTHOUGH THE LATEST QUARTERLY FINANCIAL REPORTS OF A NUMBER OF MINING COMPANIES SHOW AN INCREASE IN REVENUE COLLECTION, OVERALL PROFITABILITY REMAINS UNDER PRESSURE. ON THE ONE HAND, THE VOLATILITY OF THE VALUE OF THE ENCODED ASSET PORTFOLIO IS DELAYING PROFIT PERFORMANCE; ON THE OTHER HAND, THE CONSTRUCTION OF THE AI DATA CENTRE REQUIRES SIGNIFICANT CAPITAL EXPENDITURE, AND THE INCREASING INVESTMENT IN EQUIPMENT, SUCH AS ELECTRICITY EXPANSION, INFRASTRUCTURE DEVELOPMENT AND GPU, HAS CONTRIBUTED TO INCREASING OPERATING COSTS, LEAVING MOST MINING COMPANIES IN A STATE OF LOSS。

It is a matter of concern that, despite widespread underperformance, the equity prices of the relevant mining companies have risen significantly, meaning thatThe current focus of market attention is not on short-term profitability, but rather on the growth space of mining companies as a new generation of computing infrastructure operators。

MINE SURVIVAL HAS ESCALATED AND THE AI TRANSITION STILL NEEDS TO CROSS MULTIPLE LEVELS

The weakening of the Bitcoin market is exacerbating the environment in which mining companies survive。

According to Capriole Investments, as at 18 June, the average production cost of Bitcoin was approximately US$ 63,707, of which electricity was about US$ 50,965, with a profit margin of only 17.45 per cent for miners. In the last 30 days, the profit margin of miners has contracted by 47.8 per cent. At the same time, Luxor Hasharte Index also showed that, as at 18 June, 1 TH/s calculator-day returns had fallen to $0.032, a significant decline from the previous year's equivalent of $0.053。

Mining revenues continue to shrink, many mining companies have to sell bitcoin to maintain cash flows, pressure on small and medium-sized mining enterprises to survive has intensified, and mining resources are being concentrated to head players. Currently, the three main mining ponds, Fundry USA, AntPool and F2Pool, together, account for 59 per cent of the market share of the whole network. By contrast, in 2022, the top three bitcoin ponds combined accounted for only 44 per cent of the calculus market share。

While traditional mining operations have been depressed, the surge in demand for the AI data centre is also driving the market to reassess the value of mining companies. In its latest study, VanEck noted that the most valuable assets of mining companies are not mining machines, but electricity resources, power transformer capacity, land reserves and data centre infrastructure, which are currently the most scarce core resources in the AI industry. BecauseAI CLIENTS ARE WILLING TO PAY MUCH HIGHER ELECTRICITY PRICES AND RENTS THAN TRADITIONAL MINING OPERATIONS, AND AI INFRASTRUCTURE IS EXPECTED TO BE THE MAIN GROWTH ENGINE FOR MINING COMPANIES OVER THE NEXT DECADE。

According to the Institute Bernstein report, over $90 billion in AI infrastructure has been published by super-large cloud manufacturers, AI cloud service providers and chip companies, covering about 3.7 GW power capacity. At present, the pursuit of electricity resources is at the core of AI infrastructure competition, with Bitcoin mining companies collectively controlling more than 27 GW planned electricity capacity. In parts of the United States, a new 1GW power access cycle may last up to 50 months, making existing mines an important landing point for the expansion of the AI data centre。

However, AI transition is far from an easy path. VanEck stated that the current market was still in the early stages of the AI transition and that the enterprise valuation was mainly measured on the basis of gross electricity power available (Gross Energy Power). A higher valuation premium is generally available to mining companies that have entered into an AI lease, while projects that have stayed only in the planning phase are difficult to obtain market recognition。The future industry valuation logic will shift gradually from “power capacity” to “project delivery capacity” and eventually return to core indicators such as cash flow, capital return and tenant quality。WITH ONLY ABOUT 25 PER CENT OF CONTRACTED CAPACITY CURRENTLY DELIVERED BY THE INDUSTRY, THE TIMELY AND BUDGETED COMPLETION OF THE AI DATA CENTRE WILL BE A KEY FACTOR IN DETERMINING THE ENTERPRISE VALUATION。

VanEck also stressed thatTHE QUALITY OF AI TENANTS WILL HAVE A DIRECT IMPACT ON THE VALUATION LEVEL OF MINING COMPANIES。MEGA-SCALE CLOUD-MAKERS LIKE MICROSOFT, AMAZON, GOOGLE AND OTHERS CAN BRING MORE STABLE CASH FLOWS AND LOWER FINANCING COSTS, WHILE SMALL GPU CLOUD-SERVERS CAN RESPOND TO HIGHER OPERATING RISKS AND CAPITAL COSTS。

And the huge investment required for transformation is testing the financial power of mining companies. VanEck predicts that the transition of mining companies to AI infrastructure is still subject to huge capital expenditure demandsShort-term financing has a shortfall of about $50 billion and long-term capital needs or $221 billion。

Under enormous financial pressure, many mining companies have begun to raise funds in many ways. Minerals such as Iris Energy, TeraWul, Bitfarms and CleanSpark, for example, finance low interest rates and future swing space to attract investors through the issuance of convertible bonds, while companies such as Core Scientific, Terrawulf, Mara, Bitdeer and Riot Platforms choose to sell and even clear part of their reserves of bitcoin to sustain the transformation of AI。

In addition, a number of mining companies have begun to secure project financing support and reduce overall operating risks by locking in future revenues through long-term AI or HPC contracts. For example, CoreWeave entered into a $6 billion AI cloud service cooperation agreement with Jane Street; IREN obtained a $9.7 billion AI cloud computing contract; Hut 8 signed a $9.8 billion data centre lease agreement; and Bitdeer worked with the Norwegian DCI to build the country ' s largest AI data centre project。

FOR MINING COMPANIES, AI AT THIS STAGE UNDOUBTEDLY PROVIDES A MUCH MORE IMAGINATIVE PATH TO DEVELOPMENT THAN TRADITIONAL MINING OPERATIONS. HOWEVER, THIS TRANSITION IS NOT SIMPLY A SWITCH FROM MINING TO SALES POWER, BUT IS ESSENTIALLY A LONG-TERM COMPETITION AROUND FINANCE, RESOURCES AND IMPLEMENTATION CAPACITY。