In the current global market immersed in the frenzy of AI bull market and technology stocks continuing to lead, Arthur Hayes, the top macro expert in the cryptocurrency industry, has released a disruptive prediction. Through the four dimensions of energy cycle, American election game, global liquidity structure and AI industry bottom logic, he clearly gave the core conclusion of the future market: in the third quarter of 2026, the AI super foam that has lasted for several years will usher in a systematic collapse, driving the collective adjustment of global risk assets, and Bitcoin will follow the slump synchronously; However, after the foam is cleared, a new round of monetary easing will restart, and BTC will usher in a nirvana style boom, going out of the extreme market of falling first and rising later.

Hayes breaks away from the surface logic of single asset fluctuations and hits the core cognitive bias of the current market: although the market seems to have abundant liquidity and the AI market is booming, the vast majority of incremental US dollar liquidity has already been drained by huge capital expenditures in the AI track, which is also the core reason for Bitcoin's recent weakness and continuous underperformance compared to technology stocks. The upcoming oil price boom, AI giant IPO pressure, and the US election political game will become the triple ultimate edge to pierce the AI foam.

1、 Core underlying logic: Global incremental dollars, all absorbed by AI

In previous cycles, loose US dollar liquidity was bound to benefit cryptocurrency assets such as Bitcoin, but this round of market trend completely breaks the historical pattern. Since the landing of ChatGPT in November 2022, which ignited the AI wave, and FTX hit the bottom of Bitcoin, the market trend has shown extreme differentiation: Bitcoin has risen 7 times in three years, while AI core targets such as Nvidia have risen more than 11 times. Recently, Bitcoin has continued to decline by 50%, but AI technology stocks still maintain resilience.

Liquidity diversion is the fundamental issue. Hayes reviewed the latest macro data and found that the global M2 currency has increased by about $1.5 trillion since 2022, while the scale of new debt financing for AI data centers and computing infrastructure during the same period is exactly $1.5 trillion. This means that all the newly added US dollar liquidity in this round has been fully absorbed by the high capital intensive AI industry, and Bitcoin has no incremental funding support.

The AI industry is an extreme capital consuming industry, requiring sustained huge capital expenditures for data center construction, chip iteration, and computing power operation and maintenance. Since 2025, AI credit financing has experienced explosive growth, with a concentration of $1.3 trillion in debt landing, which corresponds to the continued weakness of Bitcoin after reaching a historical high in October 2025. In the pattern of AI siphoning off all liquidity, the overall cryptocurrency market has fallen into a state of capital depletion, with only AI racing alone.

2、 Triple deadly blade: Q3 AI foam is bound to burst

In Hayes' deduction system, the seemingly indestructible AI bull market is actually fragile. The three core risks will erupt in the third quarter of this year, completely bursting the super foam for many years.

1. Energy costs skyrocket, piercing through the foundation of AI profitability

The core essence of AI is the industrial model of "transforming energy into intelligence", with the underlying support of computing power being electricity, and the marginal cost of electricity highly dependent on fossil fuels such as oil and natural gas. The current geopolitical conflict between the United States and Iran continues to escalate, causing disruptions to shipping in the Strait of Hormuz and putting pressure on the global energy supply chain.

At present, global energy inventories have fallen to historic lows, and as conflicts continue, the energy supply gap will continue to widen. Even if the US and Iran reach a settlement in the future, countries' restocking and preventive hoarding behavior will continue to drive up oil prices.

High oil prices will directly drive up the operating costs of AI data centers, compressing the profit margins of companies such as Google, OpenAI, and Anthropic. The increase in model call costs, slowdown in commercialization growth, and decline in return on capital expenditures will completely question the narrative of AI exponential growth in the market, leading to a systematic decline in AI enterprise valuations.

2. Trillion level AI giant IPOs are concentrated and landing, and the market is unable to undertake them

Around September this year, SpaceX, OpenAI and Anthropic, the world's three largest AI giants, will launch IPO in a centralized manner. The total financing scale is far larger than the total IPO of all technology companies in the Internet era, and the supply pressure is unprecedented.

Taking SpaceX as an example, its IPO valuation is as high as $1.8 trillion, corresponding to a hundredfold sales valuation, and the initial circulating share capital is only 4% -5%, which is a typical high valuation and low current overall asset. This extreme pricing that overdraws future price increases is almost impossible to meet the market's expectations of a surge. At the same time, three trillion dollar AI companies have centralized lifting of restrictions and financing, and market liquidity is simply unable to accommodate the massive supply of new shares.

Once the IPO market falls short of expectations, the market will directly determine that the AI market has peaked, triggering a chain of stampede selling, and the valuation multiples of the AI sector will systematically shrink.

3. The game of the US presidential election has spawned a storm of "anti AI" policies

The current election situation in the United States is deadlocked, and the Republican Party is likely to lose control of the House of Representatives. Once the Democratic Party takes control of the House of Representatives, it will continue to balance the Trump camp through methods such as subpoena review and judicial accountability. To maintain his governing base, Trump will use his most effective campaign weapon: attacking AI monopolies, increasing AI regulation, imposing a windfall tax on tech giants, and suspending the disorderly expansion of data centers.

The AI industry is currently facing controversy among the public, and the social sentiment of data centers driving up inflation and AI replacing employment continues to ferment, making it a highly popular topic. Trump doesn't need to implement substantive legislation, just turning around through campaign rhetoric is enough to trigger market panic.

History has long verified the impact of political rhetoric on technology stocks: the previous public opinion conflict between Trump and Musk caused Tesla to plummet by 18% in a single day; South Korean politicians mentioned AI regulation, which directly triggered the stock index to almost hit the limit down. Trump's anti AI narrative will become the last straw to crush the AI foam at the white hot stage of the election in the third quarter of this year.

3、 Federal Reserve Policy Dilemma: Expectations of Tightening Liquidity, Exacerbating Asset Downturn

After the new Federal Reserve Chairman Kevin Walsh took office, the market's optimistic expectations for interest rate cuts continued to fall through. The current two-year US Treasury yield is 0.5% higher than the effective federal funds rate, and this key spread signal clearly indicates that the market expects sustained high inflation, and the Federal Reserve not only cannot cut interest rates, but may even raise them.

Due to the energy inflation caused by the US Iran conflict, the Federal Reserve is unable to fulfill its loose expectations. The interest rate at the interest meeting in June will remain unchanged. Against the background of high inflation, high risk and AI foam loosening, the interest rate maintenance will be interpreted by the market as a hawkish signal, further tightening global liquidity and suppressing the valuation of all risky assets.

With the sharp decline in AI stock prices, deteriorating corporate profits, and rising credit bad debts, the banking system will rapidly shrink credit, and the market will usher in a phase of credit destruction. The deep correction of global risk assets is inevitable.

4、 Bitcoin's ultimate trend: short-term market crash, super reversal in the second half of the year

For the future market of the encryption market, Hayes gave a clear layered configuration strategy and cycle prediction, and the core conclusion was very clear: there was no risk aversion attribute in the short term, and it fell sharply following the collapse of the AI foam; Long term benefits from the resumption of monetary easing, ushering in a new round of super bull market.

In the short term, during the period of global liquidity tightening, AI collapse, and market panic clearing, all risky assets are not spared. Therefore, he has taken the lead in clearing small and medium-sized counterfeit currencies such as HYPE, NEAR, WLD, and ZEC, retaining only Bitcoin and Ethereum as core bottom positions, while using derivatives to make small short positions to hedge short-term drawdown risks.

From the perspective of the end of the cycle, the financial market turbulence caused by the AI foam burst will force the US government and the Federal Reserve to restart the super easing policy and launch a new round of large-scale bailout printing. At that time, the long suppressed liquidity of the US dollar by AI will flow back into the cryptocurrency market.

Bitcoin, as the purest macro hedge asset, will completely break out of its independent market trend. With the support of loose liquidity, it will complete a nirvana style rebound and emerge from the bipolar trend of "falling first and then rising".

5、 The current optimal trading strategy

Based on the deterministic market trend in the third quarter, Hayes has released the latest holding strategy of its fund Maelstrom:

Firstly, the heavy warehouse energy sector. Regardless of the continuation or reconciliation of the US Iran conflict, the pattern of low global energy inventories and strong demand for replenishment will not change. The long-term upward trend of oil prices is clear, and the energy sector is currently the most certain safe haven and offensive track.

Secondly, liquidate non mainstream encrypted assets. Small and medium-sized currencies have no fundamental support and no liquidity to undertake, and their decline will far exceed that of Bitcoin in a systematic pullback. Priority should be given to break even and exit the market.

Thirdly, maintain core positions in BTC and ETH. Not afraid of short-term retreat, anchor the inflection point of liquidity easing in the fourth quarter, patiently go through the dark cycle of foam bursting, and wait for the super rebound market.

Fourthly, tactical hedging of risks. Using derivatives to short slightly to hedge against the risk of a systemic correction in the third quarter, while balancing principal safety and trading flexibility.

Conclusion

The core contradiction of the market in 2026 is no longer whether AI will continue to grow, but whether the foam valuation will continue, when liquidity will turn, and when policy risks will land. Arthur Hayes' cycle deduction clearly points out that the third quarter will be the key watershed of global assets, and the collapse of AI foam will trigger a round of global risk asset shuffle.

Short term panic and decline are inevitable market clearing processes, but the end of a crisis is always an opportunity. When the foam bursts and easing resumes, Bitcoin, which has been suppressed for a long time, will usher in a new round of super bull market with a higher level. The current extreme adjustment is the starting point for the new round of market momentum.