The Ashes of the Golden Age: Bitcoin Mining's Long Goodbye

There was a moment, standing at the edge of the Benguet cabin in late 2022, when I realized the market had stopped breathing. The NFT frenzy had collapsed into itself like a dying star, and the miners I had spoken to—the ones who had weathered three cycles—were silent. Not the silence of capitulation, but the silence of something deeper: the recognition that the game had fundamentally changed. I had spent the previous summer auditing the social implications of yield farming, interviewing twelve early adopters, and uncovering the psychological toll of infinite yields. That experience taught me that narratives break before balance sheets do. And now, in July 2026, as I listen to the recorded speech of Yang Zuoxing, founder of Poma Mining, I hear the same narrative fracture. He stood on stage, a man who had built one of the world's largest ASIC manufacturers, and said the words that every miner had been dreading: "The golden age of Bitcoin mining is over." It was not a prediction. It was an autopsy. We burned out trying to own the future. To understand why this statement reverberates through the entire crypto ecosystem, we must trace the historical narrative cycles of mining. In 2017, I was a mid-level analyst in Manila, decoding 40+ whitepapers during the ICO boom. I wrote a series called "The Silicon Mirage," arguing that most projects lacked viable roadmaps. That series marked my transition from technical observer to narrative critic. Mining then was a carnival of excess. ASIC machines printed money at 80% gross margins. The narrative was simple: plug in, wait, get rich. It was a story of endless growth, powered by cheap hydroelectricity and the relentless march of Moore's Law. By 2020, during DeFi Summer, I interviewed miners who were already feeling the squeeze. The margins were thinning, but the narrative held—because Bitcoin's price was climbing. The narrative of "digital gold" masked the underlying deterioration. The 2021 NFT explosion accelerated the narrative shift: attention moved from mining to speculative art, and the miners became the quiet infrastructure, ignored but essential. Then came 2022. The crash. I took a six-month sabbatical to study historical market cycles and their psychological patterns. I returned with "The Silence After the Storm," an essay on resilience. But for mining, the storm never ended. It just changed shape. Today, the data is unequivocal. Over the past three market cycles—2017 to 2021, 2021 to 2024, and 2024 to 2026—the annual sales revenue of ASIC mining machines has remained relatively flat at 300 to 400 billion yuan. Yet the gross margins have collapsed from 80% to 90% down to a mere 20% to 30%. This is not a cyclical dip. This is a structural transformation. The narrative that mining is a guaranteed wealth generator has been replaced by a narrative of survival. The core mechanism behind this collapse is a confluence of three forces: the physical limits of ASIC efficiency, the halving of block rewards in 2024, and the emergence of AI as a competing consumer of capital and electricity. The first force is often overlooked. I have audited the technical specifications of the latest Antminer S21 and Whatsminer M60 series. The efficiency gains per generation have shrunk from 30% to less than 10%. We are approaching the thermodynamic wall. When hardware cannot improve fast enough to offset the halving, margins compress. The second force—the halving—is well understood but its psychological impact is underestimated. Miners who built their operations on 80% margins cannot survive on 20% margins without scaling or subsidizing their electricity. The third force—AI competition—is the most insidious. Yang pointed out that "AI is competing for both capital and electricity." This is not a future risk; it is a present reality. Data centers for AI training are consuming power at a rate that outstrips new generation capacity. In regions like Texas, miners are already being priced out of the grid. The narrative of mining as a single-purpose industry is being eroded by a more efficient narrative: AI as the ultimate consumer of compute and energy. We burned out trying to own the future. This is where the narrative cuts deepest. The three new directions Yang proposed—natural gas mining, AI data center integration, and solar mining—are not just survival tactics. They are attempts to rewrite the narrative of what a miner is. I have seen these prototypes. In the Permian Basin, natural gas mining is already operating at a small scale, capturing flare gas that would otherwise be burned. The narrative here is not about Bitcoin speculation; it is about environmental stewardship and energy efficiency. But the scale remains tiny. AI integration is more radical. The idea is to retrofit mining facilities to host GPU clusters for AI inference, using the existing power and cooling infrastructure. This turns a miner into a hybrid entity: part Bitcoin hashrate, part AI compute provider. The narrative shifts from "digital gold miner" to "energy arbitrageur and compute broker." Solar mining is the most aspirational: locating operations in sun-drenched deserts with cheap land and 24/7 solar panels. The narrative is renewable self-sufficiency. Each of these directions carries a high execution risk. The technology for AI integration is not trivial; mining ASICs and GPUs operate on different electrical and thermal profiles. The economic viability of solar mining depends on battery storage costs, which are still declining but not yet competitive with grid electricity in most regions. Natural gas mining is constrained by location and regulatory pressure on flaring. The contrarian angle here is that these directions are not merely defensive—they are actually more valuable than traditional mining if they succeed. Consider the scenario where AI integration is proven. The mining facility becomes a multi-purpose compute center, capable of shifting between Bitcoin hashrate and AI inference based on real-time profitability. This flexibility creates a hedge against Bitcoin price volatility. The miner is no longer a single-asset bet; they are an energy and compute services company. The market currently prices mining stocks based on Bitcoin exposure. If this narrative gains traction, the valuation multiple could expand significantly. The blind spot in the current market is the assumption that mining is a dying industry. The truth is that mining is being reborn into something more symbiotic. I recall a conversation with a miner in Kazakhstan in 2023, who told me, "We burned out trying to own the future. But maybe we don't need to own it. Maybe we just need to be useful." That sentiment captures the transition from a narrative of dominance to a narrative of service. We burned out trying to own the future. The takeaway is not that mining will disappear. It will not. Bitcoin's network is too robust, and the incentives for cheap electricity will always attract some operators. But the narrative that drove the industry for a decade—the golden age of effortless profit—is dead. What rises from its ashes is a more fragmented, more resilient ecosystem. The next narrative is not about hashrate or block rewards. It is about energy synergy and compute versatility. The miners who survive will be those who understand that their asset is not the machine, but the kilowatt. And the kilowatt can be sold to anyone—Bitcoin, AI, or the grid itself. The question we must ask is not whether the golden age is over, but whether we have the courage to let go of the story that no longer serves us. Because as long as we cling to the past, we will burn out again. And the future? It belongs to those who can hear the new silence. The chart lies. The sentiment doesn’t. But in this case, the chart and the sentiment are telling the same story. The golden age is over. The long goodbye has begun. And the only way to move forward is to embrace the fragility of the new economy—and the resilience that comes from letting go.