Silence is the loudest warning. When Core Scientific’s shareholders rejected a $9 billion acquisition offer, the market exhaled—a collective sigh that masked something deeper. The deal was dead, but the questions it buried are alive: What does a mining company’s turn toward AI mean for the decentralized infrastructure we claim to build? And why, in a bull market that rewards grand narratives, did the board choose to stand alone with AMD by their side?
I have spent the last decade watching the physical layer of crypto—the steel, the power lines, the cooling towers. At 38, after auditing the governance of a dozen DAOs and mapping the liquidity architecture of DeFi, I have learned that the most important signals are often the quietest. Core Scientific’s story is not about a rejected sale; it is about the geometry of power, and how we measure trust in systems that breathe electricity.
Let me pull back the curtain on the infrastructure layer. Core Scientific is a Nasdaq-listed company (CORZ) that operates two intertwined businesses: Bitcoin mining and AI data center hosting. The latter is a new frontier, converting mining facilities into high-performance computing (HPC) clusters for AI workloads. The AMD partnership, announced alongside the rejection of the $9 billion bid, is the central piece of their strategic pivot. But as someone who has spent years dissecting the technical elegance of Ethereum’s early smart contracts, I know that a partnership announcement is not a proof of delivery. It is a promise. And promises, in a bull market, are often confused with progress.
The core of the matter lies in the transformation of energy infrastructure. Bitcoin mining is a business of converting cheap electricity into digital scarcity. AI hosting is a business of converting that same electricity into computational throughput. The two are not identical twins; they are distant cousins. A mining rig is a simple ASIC that solves SHA-256 hashes. A GPU cluster for AI requires liquid cooling, high-bandwidth networking (InfiniBand or RoCE), and a software stack that can orchestrate thousands of parallel processes. Core Scientific has experience with power procurement and facility management, but the leap into AMD’s Instinct GPU ecosystem introduces a new variable: the maturity of AMD’s ROCm software compared to Nvidia’s CUDA. From my work modeling the composability of Uniswap and Compound, I know that ecosystem lock-in is a form of gravity. Nvidia’s CUDA has a gravitational pull that AMD is still trying to escape. The partnership is a bet on that escape, and it carries the weight of a thousand unverified benchmarks.
But the deeper story is not about chips. It is about the silent architecture of trust. Core Scientific’s shareholders rejected $9 billion because they believe the company is worth more—not just in market cap, but in the intrinsic value of its power contracts. These contracts, often locked in at low rates for years, are the true asset. In a world where energy prices are volatile, a long-term power purchase agreement (PPA) is a form of arbitrage that no algorithm can replicate. The geometry of this market is simple: whoever controls the cheapest power controls the future of computation. Core Scientific is betting that their power is worth more than the easy exit.
Here is where the contrarian angle emerges. Many will celebrate the rejection as a vote of confidence in the AMD partnership. But I see a different risk: the company is moving from a decentralized service (securing the Bitcoin network) to a centralized one (serving AI workloads). The very ethos of crypto is challenged when our infrastructure becomes a tool for large language models trained behind corporate firewalls. The Bitcoin network is a public good; an AI data center is a private utility. The transition is not evil, but it is a pivot away from the original vision of permissionless, borderless computation. We must ask: are we building the cathedral, or are we selling the stones?
Based on my experience auditing the governance tokens of mid-sized DAOs during the 2022 bear market, I learned that the most dangerous flaws are the ones that look like features. The AMD partnership looks like a feature—a diversification of revenue, a hedge against the halving. But it is also a concentration of risk. If AMD’s supply chain falters, or if the Instinct GPUs fail to match Nvidia’s performance, Core Scientific will have committed capital to a path that may not deliver. The silence from the company on technical details—no delivered megawatts, no utilization rates, no benchmark results—is the loudest warning. In the 2020 DeFi summer, we saw protocols raise millions without a working product. The same pattern is emerging in the infrastructure layer, but with real power plants and real chips.
Prune the dead branches, save the tree. This is a principle I have applied to every system I analyze. Core Scientific’s tree has two branches: mining and AI hosting. The mining branch is mature, generating cash flow. The AI branch is an experiment. The rejection of the $9 billion sale is a decision to let the experiment grow, but it also means that the tree must bear fruit quickly. The market is watching. The geometry of trust is not forgiving.
Let me offer a forward-looking thought. The convergence of crypto infrastructure and AI is inevitable, but it will test our values. Will we allow these facilities to become the backbone of a centralized AI oligopoly, or will we design them to serve the public good? The answer lies not in the boardroom, but in the code we write and the contracts we sign. Core Scientific’s story is a mirror for the entire industry: we are at a crossroads where power is both literal and metaphorical. The choices we make today will echo through the next decade.
As I write this from Beijing, watching the sun rise over a city that breathes data, I am reminded that the most beautiful systems are those that remain true to their purpose. Geometry remembers what markets forget. The market may forget the terms of this rejected deal, but the geometry of energy, trust, and human intent will persist. The question is not whether Core Scientific’s gamble pays off, but whether we, as a community, remember why we started building in the first place.