The shareholders of Core Scientific voted to reject a $9 billion acquisition offer. The canvas shifted, but the buyer remained—a ghost of a deal that now hangs over the company's next move. The rejection came alongside the announcement of a partnership with AMD, a narrative pivot from Bitcoin mining to AI infrastructure. But as I traced the ghost of the 2017 token sale audits, I recognized a pattern: strategic announcements often mask the absence of technical substance. The vote was a collective bet that the company's stock, currently trading at a fraction of that implied valuation, could climb higher on the back of an AI pivot. But the market's euphoria may be mistaking a press release for a proof-of-work.
Context: Core Scientific emerged from Chapter 11 bankruptcy in early 2024, a phoenix from the ashes of the crypto winter. The company operates some of the largest Bitcoin mining facilities in North America, with a fleet of ASICs and access to cheap, stranded power through long-term power purchase agreements. In 2023, they began a transformation: retrofitting mining sites to host high-performance computing (HPC) and AI workloads. They signed multi-year contracts with CoreWeave, a pure-play AI cloud provider, to lease out GPU capacity. The AMD partnership, announced in late 2024, is the next chapter in this narrative. AMD would supply Instinct GPUs to Core Scientific, ostensibly to power the AI hosting business. The $9 billion acquisition offer—whispered to be from a consortium of infrastructure investors—was rejected by shareholders, who argued that the AMD deal alone justified a higher valuation. The market response was immediate: CORZ stock surged 15% on the news. But the surge was a narrative event, not a technical one.
Core Insight: The Narrative Velocity of AI-Mining Convergence
Mapping the invisible liquidity flows of summer 2020, I saw how DeFi protocols gained value not from code audits but from community sentiment. Core Scientific's current situation is a mirror: the AMD partnership is a narrative accelerant, but the underlying infrastructure is still being built. The company's core advantage is its power infrastructure—cheap, stranded energy from hydro, solar, and natural gas plants that were once dedicated to Bitcoin mining. Converting these sites to AI data centers is engineeringly feasible, but it requires solving a different set of problems: liquid cooling, high-density racking, InfiniBand networking, and GPU cluster orchestration. AMD's ROCm software stack is still catching up to Nvidia's CUDA; any production-scale deployment will face integration headaches. Based on my experience auditing AI infrastructure projects during the 2021 NFT pivot, I've seen that partnerships without technical milestones are often little more than optionality. The AMD deal disclosed no committed capacity, no minimum purchase quantities, no revenue-sharing structure. It's a supply agreement, not a revenue guarantee.
Every codebase is a whispered promise; every contract is a loud one. The shareholders who rejected the $9 billion offer are betting that the AMD partnership will unlock value exceeding that amount. But the numbers don't add up without a clear path to revenue. Core Scientific's mining business is volatile—post-halving, the block reward is 3.125 BTC per block, and the company's hashrate is a fraction of the network. The AI hosting business, meanwhile, is nascent. CoreWeave contracts provide some revenue visibility, but the market is pricing in a full-scale pivot. The narrative velocity is high, but the narrative durability is low. I recall from my 2017 token sale audit sprint that projects with the most compelling vision statements often had the fewest working products. The same heuristic applies here: the AMD partnership is a vision statement, not a balance sheet.
Contrarian Angle: The Ghost of the $9 Billion Offer
The contrarian view is that the shareholders may have made a mistake. The $9 billion offer likely represented a full valuation of the company's assets—both mining and potential AI capacity. Rejecting it implies that management believes they can create more value, but the risks are substantial. The AI infrastructure market is crowded with incumbents like AWS, Azure, and Google Cloud, as well as specialized players like CoreWeave and Lambda. Core Scientific's advantage—cheap power—is not unique; many mining companies are pivoting to AI. The AMD partnership, while diversifying chip supply, does not solve the fundamental challenge of customer acquisition. The market may be overestimating the speed of the pivot. Summer taught us that liquidity has a heartbeat; now, narrative has a velocity. But velocity without direction is just noise. The $9 billion offer was a concrete valuation; the AMD partnership is an abstract promise. The market's enthusiasm may be pricing in a future that never materializes.
Furthermore, the shareholder rejection may be a case of narrative overconfidence. The board's decision to announce the AMD partnership alongside the rejection was a strategic move to frame the AMD deal as a value-creating event. But the market may be misreading the signal. Based on my 2022 bear market sentiment reconstruction, I found that companies that successfully pivoted their messaging to align with regulatory or technological trends often saw short-term stock gains, but long-term value depended on actual execution. Core Scientific's pivot is a narrative shift, not a business shift. The company still has billions in debt from its bankruptcy restructuring, and any AI infrastructure build-out requires significant capital expenditure—likely through equity dilution or further debt. The $9 billion offer may have been the best exit for shareholders.
Takeaway: The Next Narrative to Watch
The next narrative to watch is the delivery of AI capacity. If Core Scientific can announce a specific number of megawatts deployed with AMD GPUs, with utilization rates above 80%, the narrative will solidify. If they delay, or if the ROCm stack proves problematic, the stock will retreat. The ghost of the $9 billion offer will remain as a benchmark—a constant reminder of the value that was rejected. The canvas shifted, but the buyer remained a phantom. Will the company paint a masterpiece, or will the canvas tear under the weight of unrealized promises?