The Bitcoin Abstention | Stellae Liquiditas Research

A position statement of Stellae Liquiditas LLC · Division 03 · Digital Assets

Stellae Liquiditas is a principal-only digital asset treasury built on conviction, not consensus. We accumulate the settlement networks we believe will carry institutional value across the next several decades: XRP, XDC, XLM, HBAR, ADA, and CRO. Bitcoin is not among them, and its absence is deliberate.

This is not a trade or a timing call. It is a structural judgment. We hold that Bitcoin carries three compounding liabilities — a cryptographic exposure it cannot patch quickly enough, a governance model too slow to adapt, and an energy profile increasingly antithetical to the direction the entire economy is now moving. Below is the whole of our reasoning, on the record. Every factual claim is anchored to a peer-reviewed paper, an official white paper, a protocol proposal, or a first-party institutional announcement — the Sources are listed at the end.

I. The Quantum Exposure

Sealed March 30, 2026

Bitcoin’s signatures rest on elliptic-curve cryptography (ECDSA). That scheme is broken by Shor’s algorithm the moment a sufficiently powerful, fault-tolerant quantum computer exists. This is not speculation; it is arithmetic on a known clock.

The exposure is already on-chain and permanent. A peer-reviewed on-chain study by researchers at Deloitte Netherlands found that on the order of four million BTC — roughly a quarter of the coins in circulation at the time — sat in address formats that directly expose the public key (early Pay-to-Public-Key outputs and reused Pay-to-Public-Key-Hash addresses). Among these, the earliest Satoshi-era P2PK coins — on the order of 1.1 million BTC identified through the Patoshi on-chain pattern — have had their public keys visible in the clear since 2009 and cannot be hidden retroactively. Once a public key is exposed, that coin is a standing target, and the “harvest now, decrypt later” posture means an adversary can record those keys today and act on them the day the hardware matures.

And the resource cost of that attack is collapsing. In 2026, Google Quantum AI published a white paper pointedly titled Securing Elliptic Curve Cryptocurrencies against Quantum Vulnerabilities, estimating that the P-256 curve underlying Bitcoin could be attacked with fewer than 1,200 logical qubits — on the order of a few hundred thousand physical qubits — potentially completing in a matter of hours, roughly a twentyfold reduction from the group’s earlier estimates. A parallel, AI-assisted resource estimate from a Caltech-linked team (Oratomic) shortened the picture further. In direct response, Cloudflare — which secures a large share of global internet traffic — accelerated its own post-quantum migration deadline to 2029.

On the hardware side, the last major bottleneck to scale was cleared in early 2026. In a paper published in Nature, a Stanford team led by Adam Shaw and Jonathan Simon introduced a “cavity-array microscope” — single atoms housed in micro-optical cavities with integrated lenses — achieving fast, non-destructive, parallel readout of every qubit at once, across an array of more than 40 cavities and a prototype exceeding 500. Reading qubits in parallel rather than one at a time is the step the field had been stuck behind on the path toward million-qubit machines.

Read together, these are not separate headlines. They are the numerator and denominator of the same fraction closing: the cost to break the cipher is falling while the machines to run the attack are being built. Bitcoin’s exposed coins sit directly in that path.

The deeper problem is not the threat — it is the response. Bitcoin’s fix requires network-wide migration to post-quantum signatures, and the community is openly split over how. BIP-360 proposes quantum-resistant output types; BIP-361 contemplates freezing vulnerable coins to prevent theft — a proposal that fractured the ecosystem, because freezing Satoshi-era holdings raises a governance question more explosive than the cryptography itself. Bitcoin’s history says this will not move quickly (Section II).

Contrast this with where our capital actually sits. The XRP Ledger supports native, account-level key rotation in its base protocol — an account can replace a compromised signing key without moving funds or changing addresses, a structural capability Bitcoin lacks. Ripple has published a post-quantum roadmap targeting a fully quantum-resistant XRPL by 2028, testing NIST-standard ML-DSA signatures in partnership with the post-quantum firm Project Eleven. When your recovery mechanism is “ask millions of self-custodians to voluntarily migrate before an unknown deadline,” you do not have a recovery mechanism. You have a hope.

II. Governance Ossification: The Asset That Cannot Move

Sealed March 27, 2025

Bitcoin’s defenders praise its resistance to change. We read that same trait as an inability to respond to existential risk on a survivable timeline.

The protocol record is unambiguous. Bitcoin’s Segregated Witness upgrade activated in 2017. Its next consensus upgrade, Taproot (BIPs 340–342), was first proposed in 2018 and did not activate until November 2021, at block 709,632 — Bitcoin’s first protocol upgrade in over four years — after which adoption across wallets and services still unfolded over subsequent years because upgrades are opt-in. These were relatively uncontroversial improvements, and they still crawled through a miner-signaling process that hands effective veto power to a fragmented set of participants.

Now apply that machine to an emergency: a network that needed the better part of a decade to adopt a signature optimization, negotiating a full post-quantum migration against hardware progress that just compressed twentyfold in a single research cycle. The two clocks are not compatible.

Purpose-built settlement networks move on a different order of magnitude. Per the XRP Ledger’s documentation, an amendment reaches the ledger once it holds 80% validator support sustained over a two-week window — governance measured in weeks, not years — and the ledger already supports account-level key rotation natively. Agility is not a luxury here. In a decade defined by post-quantum transition, tokenization standards, and regulatory integration, it is the whole game. We will not anchor a multi-decade treasury to an asset that cannot patch itself in time.

III. The Green-Energy Antithesis

Sealed December 17, 2025

This is where our conviction is strongest.

Bitcoin’s proof-of-work consensus consumes, by the Cambridge Centre for Alternative Finance’s estimate, on the order of 150 terawatt-hours of electricity per year — roughly half a percent of all electricity generated on Earth — to secure a single network. That expenditure is not incidental to Bitcoin; it is Bitcoin. The energy burn is the security model, and it cannot be optimized away without abandoning the design.

We do not accept the “but it uses renewables” defense as sufficient. Even with the Cambridge Centre estimating the sustainable share of the mining mix at roughly half, every megawatt a miner draws — clean or otherwise — is a megawatt bid away from a grid the rest of the economy now needs with unprecedented urgency.

Because the demand curve has changed shape. Per the International Energy Agency, global data-centre electricity use was on the order of 415 TWh in 2024 and is projected to roughly double to about 945 TWh by 2030, with AI-specific load growing fastest; the IEA has noted that data centres, AI, and crypto combined could exceed 1,000 TWh within this window. And that is before the coming wave of IoT, edge inference, robotics, and autonomous systems that will draw on the same finite grid.

Here is the question we keep returning to, and the one proof-of-work cannot answer: for what? An AI cluster trains models, serves billions of inferences, and powers enterprise software, scientific computing, and the autonomous infrastructure being built right now. A proof-of-work rig performs quintillions of deliberately discarded hash operations to win a lottery. When firm power becomes the scarcest input in the economy — and it is becoming exactly that — burning it on manufactured difficulty is not a neutral choice. It is a misallocation of a resource the future needs.

As a firm, we also observe the shift in where environmental attention is directed. A decade ago the energy debate fixated on Bitcoin; today the far larger and faster-growing load is AI compute, as the IEA’s own figures make plain. The scrutiny moved on. The megawatts did not — Bitcoin’s draw has not fallen, and the aggregate compute burden is climbing to record highs. We state plainly that proof-of-work’s claim on that power is the weakest of any major user.

The China Case: A Nation-State Test of the Thesis

During a strategic digital-asset tour of Hong Kong in late 2025, the firm’s ownership encountered renewed regional coverage of Hong Kong’s own regulated virtual-asset framework — positioned in deliberate contrast to the mainland prohibition — which prompted us to formalize the underlying mainland case study on the record here.

China supplies the clearest test of our thesis at nation-state scale. As of 2020, the Cambridge Centre for Alternative Finance estimated that mainland China hosted on the order of 65% of global Bitcoin hashrate — the country was, for practical purposes, the security infrastructure of the network. Starting in May 2021, the State Council’s Financial Stability and Development Committee publicly called for a halt to Bitcoin mining and trading. By September 3, 2021, eleven central government agencies — led by the National Development and Reform Commission — had formalized a nationwide mining shutdown, explicitly grouping virtual-currency mining alongside steel, cement, and other high-energy-consumption industries incompatible with the country’s carbon-neutrality targets. On September 15, 2021, the People’s Bank of China, jointly with nine other government bodies, extended the prohibition to all cryptocurrency trading and related services (publicly posted September 24, 2021), stating in the accompanying notice that overseas exchanges serving mainland residents were themselves engaged in illegal financial activity. The world’s second-largest economy, having spent years running the majority of Bitcoin’s proof-of-work security, decided it did not want it.

This is our own thesis, arrived at independently, confirmed at scale: proof-of-work’s power draw is structurally incompatible with a serious carbon-neutrality mandate.

What happened next is the more important data point — and it is the part of the public record that closes any remaining “but renewables” defense. Displaced hashrate did not disappear; it relocated, principally to Kazakhstan and the United States. A peer-reviewed study by de Vries, Gallersdörfer, Klaaßen and Stoll (Revisiting Bitcoin’s Carbon Footprint, Joule, 2022) documented the aftermath. The share of renewable electricity powering the Bitcoin network fell from 41.6% in 2020 to 25.1% by August 2021. Chinese hydropower — which miners had accessed seasonally in Sichuan and Yunnan during the wet months — was replaced by hard coal in Kazakhstan (higher-emitting per kilowatt-hour than the Chinese coal it displaced) and by natural gas in the United States, whose share in the mining mix roughly doubled from 15% to 30.8%. Estimated network emissions climbed to 65.4 megatons of CO2 annually — comparable to country-level emissions in Greece. The largest miner in the world walked away entirely, and proof-of-work’s climate footprint got worse, not better. That is not a migration accident; it is the structural signature of a security model whose only lever is more electricity, wherever it can be found. When one jurisdiction removes the cheapest power, the network purchases the next-cheapest power available — which, at the frontier of cost, is almost always the dirtiest.

Electricity is not the only environmental line item. A separate peer-reviewed study (de Vries & Stoll, Resources, Conservation and Recycling, 2021) estimated that Bitcoin’s specialized mining hardware — application-specific integrated circuits with no secondary use, obsoleted on a roughly 1.5-year replacement cycle as newer ASICs outcompete them — generated on the order of 30.7 kilotons of electronic waste per year, comparable to the small-IT-equipment waste of a mid-sized European country. AI accelerators, by contrast, retain resale value across training, inference, and scientific-computing workloads long after their frontier use ends. Same silicon budget, radically different second life.

Once the largest jurisdiction on Earth arrived at the same conclusion we did, and the network’s response was to relocate its emissions rather than reduce them, the constructive question is no longer whether proof-of-work belongs in the future power mix. It is what to do with the physical infrastructure it leaves behind. Section V takes up that question.

IV. The El Salvador Experiment: What Happens When a Nation Bets on Bitcoin

Sealed January 29, 2025

In September 2021, El Salvador became the first country in the world to adopt Bitcoin as legal tender, requiring all businesses to accept it alongside the U.S. dollar. The experiment was framed as a revolution in financial inclusion — a way to cut remittance fees for a country where foreign transfers account for roughly a quarter of GDP. Three years later, the experiment has largely collapsed under the weight of its own assumptions.

The Infrastructure Failure

On launch day — September 7, 2021 — the government’s official Bitcoin wallet, Chivo, went offline within hours. Servers could not handle demand. President Bukele acknowledged on social media that the rollout had been rushed, admitting that launching everything in three months was too much of a challenge. The wallet was pulled from app stores and new registrations were suspended. For four consecutive days, users reported an inability to download the app, access funds, withdraw from ATMs, or receive the $30 sign-up bonus the government had promised. Some users reported unexplained losses of funds. Others discovered that someone had already claimed their $30 using their identity — a consequence of inadequate verification controls. Users described the app’s functionality during that period as almost zero. The wallet was declared 95 percent fixed a week after launch. It never fully recovered public trust.

The Adoption Collapse

The $30 sign-up incentive drove initial downloads, but usage collapsed almost immediately. A National Bureau of Economic Research study found that by mid-2022, more than 60 percent of users who received the incentive had never made a single transaction. A 2023 survey by the Central American University (UCA) found that 88 percent of Salvadorans had not used Bitcoin that year. By 2024, that figure rose to 92 percent. Only 8.1 percent of the population reported using Bitcoin for any payment in 2024 — down from 25.7 percent in the incentive-driven first year. An El Salvador Chamber of Commerce survey found that 86 percent of businesses had never completed a Bitcoin sale. The unbanked population the law was designed to reach largely did not adopt it. Consumers who did use the app regularly were not the unbanked originally targeted by the Bitcoin Law. Virtually no one installed the Chivo wallet on their phone in 2022 or after. Businesses were not keeping the Bitcoin they received — they converted immediately to dollars.

The Remittance Failure

The stated purpose of the Bitcoin Law was to reduce remittance costs. Bitcoin-based remittances through Chivo totaled $39.4 million in the first four months of 2022, fell to $32 million in the same period of 2023, and declined further to $28.8 million in the first four months of 2024 — barely one percent of the $2.64 billion in total remittances over that span. Each year, the Bitcoin share shrank while total remittance volume grew. The technology that was supposed to revolutionize remittances captured a smaller share of them every year it was in operation.

The Sovereign Cost

Bitcoin’s price fell more than 50 percent from its November 2021 peak to its mid-2022 trough. Because El Salvador’s government was an active buyer — President Bukele publicly announced purchases, including 150 coins bought during a crash on the very day the law took effect — the nation’s balance sheet absorbed the full volatility of a speculative asset. Credit rating agencies cited Bitcoin exposure as a factor in downgrading El Salvador’s sovereign debt. On December 1, 2024, President Bukele himself acknowledged that the Bitcoin Law had been the most unpopular measure his government had taken.

The Unwinding

In December 2024, El Salvador reached a $1.4 billion loan agreement with the International Monetary Fund. The terms required the government to make Bitcoin acceptance voluntary for the private sector, cease accepting tax payments in Bitcoin, and wind down or sell the Chivo wallet. In March 2025, El Salvador’s legislature voted 55-2 to amend the Bitcoin Law accordingly. The Chivo wallet — the centerpiece of the experiment — is being sold or shut down. Bitcoin remains nominally legal tender, but the infrastructure built to make it function as one has been dismantled by the government that created it.

What El Salvador Proved

The El Salvador experiment is the most complete real-world test of Bitcoin as a national settlement and payment layer. It failed on every metric it set for itself. The wallet infrastructure could not handle launch-day demand. The population rejected it — 92 percent non-usage by the final year. The remittance thesis produced a shrinking share of a growing market. The sovereign balance sheet absorbed speculative losses that contributed to debt downgrades. And the experiment ended with the government agreeing to unwind it as a condition of international lending.

This is not a criticism of El Salvador’s intent. Financial inclusion is an urgent and legitimate goal. But the instrument chosen — a volatile, energy-intensive, throughput-constrained proof-of-work network — was architecturally unsuited to the task. A network that processes seven transactions per second, charges fees that spike unpredictably during congestion, and subjects every holder to the full volatility of a speculative asset is not a settlement rail for a remittance-dependent economy. It is a speculative vehicle dressed as infrastructure.

The rails that can actually serve that purpose — fast-settlement, low-energy, fee-stable networks designed for cross-border value transfer — are precisely the ones Stellae Liquiditas holds. The lesson of El Salvador is not that digital asset settlement failed. It is that the wrong digital asset was chosen.

V. Our Position: Repurpose, Not Dual-Use

Stellae Liquiditas holds, as a matter of doctrine, that proof-of-work mining infrastructure — the power contracts, substations, cooling, and buildings — should be repurposed for AI and high-performance computing as rapidly as possible.

The industry has begun this migration on its own economics. Core Scientific announced a multi-billion-dollar, multi-year arrangement to host CoreWeave’s AI compute; IREN announced a $9.7 billion agreement with Microsoft to deploy tens of thousands of next-generation NVIDIA GPUs. These are first-party corporate commitments, not forecasts — mining operators converting power and facilities to computation that produces output.

But we go further than the market’s current hedge. We reject the dual-use model — running facilities part-time as miners and part-time as AI hosts, keeping proof-of-work alive as a fallback. Dual-use is a half-measure that treats an obsolete workload as a legitimate claimant on scarce power. The correct end state is full conversion: this infrastructure allocated wholesale to computation that produces something. Not a hedge. A handoff.

VI. The Constructive Alternative

Abstention from Bitcoin is not a negative posture. It is the direct expression of what we do hold.

Every core asset on the Stellae Liquiditas balance sheet was engineered for the world described above — near-zero-energy settlement, cryptographic agility, and purpose-built utility. XRP, XDC, and HBAR finalize transactions for a rounding error of Bitcoin’s energy per settlement. XRPL can rotate keys and adopt post-quantum signatures on a survivable timeline. XDC is built for trade finance and tokenization; Stellar for cross-border settlement; Hedera for enterprise-grade throughput. These are networks that feed the compute economy rather than competing with it for the plug.

That is the whole thesis in one line: we allocate to the rails the future runs on, and we decline to secure a monument to the resource the future can least afford to waste.

Stellae Liquiditas holds no Bitcoin. It never has. And on the reasoning above, it does not intend to.

Addendum I — Market-Structure Update

The Reflexive Trap: Filed as of July 17, 2026

Sections I through VI above were written as a structural argument. What follows is the empirical record confirming that the structure has begun to fail publicly and on the tape. No claims below are forward-looking. They are reported figures drawn from SEC filings, first-party exchange disclosures, and institutional flow data.

The model required two conditions to hold simultaneously. Corporate Bitcoin treasury companies — the category pioneered by Strategy (formerly MicroStrategy) and rapidly replicated across roughly 199 public companies holding an estimated 1.264 million BTC by June 22, 2026 — were built on a compounding flywheel: shares trading at a premium to the underlying Bitcoin holdings (a metric the industry calls mNAV, or market-to-net-asset-value), which permitted management to issue new equity, use the proceeds to buy more Bitcoin, and thereby increase Bitcoin per share. That accretion attracted more premium, which funded more issuance. The flywheel required two inputs to keep spinning: Bitcoin’s price continuing to rise, and the company’s shares continuing to trade above the value of the coins on the balance sheet. When one input fails, the flywheel reverses. When both fail simultaneously, the liabilities remain while the mechanism that was meant to service them stops working.

The rally was a liquidity trade, not a fundamental re-rating. CF Benchmarks, the Financial Conduct Authority-regulated index provider and Kraken subsidiary, published data as of February 1, 2026 documenting that Bitcoin’s rolling four-year R-squared to global M2 money supply had ranged between 0.71 and 0.90 across historical cycles — among the highest macro correlations of any major asset class. The same analysis documented that this relationship broke sharply in Q4 2025: global M2 expanded approximately 12% while Bitcoin declined approximately 12% over the trailing twelve months, one of the most extreme dislocations between price and the liquidity model in the dataset’s history. The implication is not that Bitcoin is cheap. It is that the prior rally tracked monetary expansion rather than adoption or utility, and that when the liquidity tide receded, price followed. Bitcoin reached an all-time high of approximately $126,198 on October 14, 2025 — the peak of the post-ETF, post-halving liquidity cycle — and by June 2026 had corrected to a year-to-date low of $58,190, a drawdown of approximately 54% from that high.

Institutional capital is rotating out of the ETF wrapper. Spot Bitcoin ETFs launched in the United States in January 2024 and became the fastest-growing ETF category in history by assets. That inflow dynamic reversed in 2026. Per SoSoValue and Farside Investors flow data: U.S. spot Bitcoin ETFs recorded $4.06 to $4.5 billion in net outflows during June 2026 alone — the worst calendar month on record since launch, surpassing the prior record of $3.56 billion set in February 2025. BlackRock’s iShares Bitcoin Trust (IBIT), the dominant product by assets, accounted for approximately 75–79% of June’s total redemptions. Between late June and July 2, 2026, IBIT shed 35,980 BTC (approximately $2.24 billion) across ten consecutive trading days, the longest single-fund outflow streak on record for any U.S. spot Bitcoin ETF. 2026 is the first calendar year in which net flows for U.S. spot Bitcoin ETFs have turned negative in aggregate since the products launched. No Bitcoin-specific fundamental break drove the selling — macro triggers identified in contemporaneous disclosures included reaccelerating CPI and the Federal Reserve holding rates at 3.5–3.75%.

Treasury-company premiums have collapsed. At its peak in April 2025, the average mNAV across publicly traded Bitcoin treasury companies was 3.76x — investors paying $3.76 for every $1 of Bitcoin held. By June 2026, that premium had disappeared for the majority of operators. K33 Research data showed 26 of 168 publicly traded Bitcoin-holding firms trading below 1.0x NAV as of October 2025. KindlyMD/Nakamoto Holdings (NAKA), which had briefly reached 75x mNAV, collapsed 96% from its peak to 0.7x. Semler Scientific shares fell approximately 74% year-to-date 2026. Strategy shares fell approximately 52% year-to-date by mid-2026 and approximately 75% year-over-year at trough. Strategy itself had traded at 3.89x mNAV at its November 2024 peak of $543 per share; by the time the capital restructuring below was announced, that premium had compressed to a level at which conventional share issuance was no longer accretive. A sub-1.0x mNAV means a company buying one dollar of Bitcoin by issuing one dollar of new equity is destroying shareholder value on day one. The flywheel does not merely slow; it runs backward.

Strategy’s capital framework and the public record on selling. For approximately five years, Strategy’s executive chairman maintained a consistent public position that the company would never sell Bitcoin — framing Bitcoin as a generational treasury asset to be borrowed against, not liquidated. In February 2026, pressed on what the company would do if Bitcoin declined and stayed down, he stated on CNBC that Strategy would continue buying and would refinance debt before it touched its Bitcoin holdings. Four months later, the company sold 32 BTC at an average price of $77,135 — its first sale since 2022 — to fund a preferred dividend. On June 11, 2026, on stage at the BTC Prague conference, the position was revised publicly: the “never sell” framing had been intended for individual investors; the company, as a publicly traded entity with fixed obligations, retained and had always retained the right to sell when its finances required it.

On June 29, 2026, Strategy filed an 8-K with the U.S. Securities and Exchange Commission formally codifying this. The filing announced a five-part Digital Credit Capital Framework with the following disclosed parameters: a USD Reserve Policy establishing a cash reserve of $2.55 billion as of June 28, 2026 (covering approximately 17.4 months of preferred dividends and interest, against a board-mandated floor of 12 months); a revised STRC preferred stock dividend raised to 12.00% per annum effective July 1, 2026; a Digital Credit Securities Repurchase Program of up to $1.0 billion; a Class A Common Stock Repurchase Program of up to $1.0 billion; and a Bitcoin Monetization Program authorizing management to sell up to $1.25 billion of Bitcoin to supplement the USD reserve, service preferred dividends and interest, and fund repurchases. Strategy’s CEO Phong Le described the shift as moving “from one-way capital issuance to active capital management.” The company held 847,363 BTC at an aggregate cost of $64.10 billion ($75,651 average per coin) as of the filing date — meaning the average cost basis per coin exceeds the price at which Bitcoin was trading when the monetization program was authorized. The filing does not obligate any sale, but its existence — and the SEC’s public record of it — establishes that Bitcoin monetization is now a standing operational tool rather than an exceptional measure. This was confirmed by Strategy’s own CEO in the same filing. The authorization is a first-party institutional document, not a news account.

What this means structurally. The architecture that failed is not difficult to describe. Fixed, non-callable cash obligations — preferred dividends, convertible note interest, covenant maintenance — were layered on top of a volatile, non-yielding asset. The model was solvent under one condition: continuous appreciation. When appreciation stopped, the leverage that accelerated the upside began extracting in the opposite direction, and the equity issuance engine that had funded accumulation reversed into dilution. This is not a criticism of any operator’s judgment or execution. It is a description of what fixed liabilities layered on a non-yielding volatile asset do when the price assumption that justified them does not hold.

Stellae Liquiditas carries none of this architecture. The entity holds no preferred equity, issues no dividends, carries no margin financing, and has no mNAV dependency — because it is not a public company using equity issuance to lever a Bitcoin position. It is a principal-only corporate digital asset treasury, sole-member owned, with no fixed obligations owed to outside capital. The assets on the balance sheet produce no yield and are not expected to. They are long-horizon conviction positions held without leverage, without a flywheel, and without a liquidation trigger. The absence of Bitcoin from that balance sheet and the absence of the reflexive treasury model from that structure are not two decisions. They are one.

Sources

Deloitte Netherlands — on-chain study of quantum-vulnerable Bitcoin (Pay-to-Public-Key and reused addresses), arXiv, 2023.
S. D. Lerner — Patoshi pattern analysis identifying early Satoshi-era mining holdings (on-chain research).
Google Quantum AI — Securing Elliptic Curve Cryptocurrencies against Quantum Vulnerabilities: Resource Estimates and Mitigations (white paper / arXiv 2603.28846), 2026.
Oratomic / Caltech-affiliated team — AI-assisted quantum resource-estimate preprint (arXiv 2603.28627), 2026.
Cloudflare — first-party announcement accelerating its post-quantum migration deadline to 2029, 2026.
Shaw, A. L., Soper, A., Shadmany, D., … Schuster, D. I., & Simon, J. — A cavity-array microscope for parallel single-atom interfacing, Nature, 2026, DOI 10.1038/s41586-025-10035-9 (preprint arXiv 2506.10919); Stanford Report, “Light-based platform sets the stage for quantum supercomputers,” Jan 2026.
Bitcoin Improvement Proposals — BIP-360 (quantum-resistant output types) and BIP-361 (migration / vulnerable-coin handling), Bitcoin BIPs repository.
Bitcoin protocol history — SegWit activation (2017); Taproot (BIPs 340–342) activated at block 709,632, November 2021.
XRP Ledger documentation — account key rotation (regular/master keys) and the amendment process (80% validator support over two weeks).
Ripple — published post-quantum roadmap (NIST ML-DSA testing; partnership with Project Eleven; 2028 target).
Cambridge Centre for Alternative Finance — Cambridge Bitcoin Electricity Consumption Index (network electricity estimate and sustainable-energy share); Cambridge Bitcoin Mining Map (country-level hashrate distribution; China ≈ 65% share, 2020 baseline).
International Energy Agency — data-centre and AI electricity projections (≈415 TWh in 2024; ≈945 TWh by 2030; combined data-centre/AI/crypto load potentially exceeding 1,000 TWh).
State Council of the People’s Republic of China, Financial Stability and Development Committee — public statement calling for cessation of Bitcoin mining and trading activity, May 2021.
National Development and Reform Commission of the People’s Republic of China, jointly with ten other central agencies — Notice on the Regulation of Virtual Currency “Mining” Activities, September 3, 2021.
People’s Bank of China, jointly with nine other government bodies — Notice on Further Prevention and Disposal of the Risk of Speculation in Virtual Currency Trading, effective September 15, 2021 (posted on the PBOC website September 24, 2021).
de Vries, A., Gallersdörfer, U., Klaaßen, L., & Stoll, C. — Revisiting Bitcoin’s Carbon Footprint, Joule, 6(3), 498–502, 2022, DOI 10.1016/j.joule.2022.02.005.
de Vries, A., & Stoll, C. — Bitcoin’s Growing E-Waste Problem, Resources, Conservation and Recycling, 175, 105901, 2021, DOI 10.1016/j.resconrec.2021.105901.
Company announcements — Core Scientific / CoreWeave AI-hosting arrangement; IREN / Microsoft GPU deployment ($9.7 billion).
CF Benchmarks — The M2-Bitcoin Relationship: What the Data Actually Shows, CF Benchmarks Research, February 1, 2026 (R-squared 0.71–0.90 to global M2; Q4 2025 decoupling; M2-implied fair value vs. actual price gap analysis).
SoSoValue — U.S. Spot Bitcoin ETF daily and monthly net flow data (June 2026 record outflows; 2026 first year of net-negative annual flows since January 2024 launch).
Farside Investors — U.S. Bitcoin ETF flow tracker (BlackRock IBIT net outflows; May–June 2026 streak data; 35,980 BTC / ~$2.24B IBIT 10-day streak, June 22 – July 2, 2026).
K33 Research — Bitcoin treasury company mNAV tracker (26 of 168 public firms below 1.0x NAV, October 2025; average sector mNAV 3.76x April 2025 → compression through 2026).
BitcoinTreasuries.net — Public company Bitcoin holdings tracker (199 public companies, 1.264 million BTC, ~$79 billion, as of June 22, 2026).
Strategy Inc. (NASDAQ: MSTR) — Form 8-K filed with the U.S. Securities and Exchange Commission, June 29, 2026: Digital Credit Capital Framework (Bitcoin Monetization Program up to $1.25B; USD Reserve Policy; STRC dividend revised to 12.00%; holdings 847,363 BTC at $64.10B aggregate cost / $75,651 average; CEO Phong Le statement on transition to “active capital management”). Available: SEC EDGAR, accession on file.
Strategy Inc. — CNBC/Andrew Ross Sorkin interview, February 2026 (executive chairman statement on refinancing debt before selling Bitcoin).
BTC Prague 2026 conference — Executive chairman public remarks, June 11, 2026 (distinction between personal-investor “never sell” advice and company operational flexibility).
Strategy Inc. — Sale of 32 BTC at average $77,135, disclosed May 2026 (first Bitcoin sale since 2022; characterized as ad-hoc in subsequent 8-K).
President Nayib Bukele — Twitter/X statements, September 7–14, 2021 (launch-day acknowledgment; “95% fixed” admission).
National Bureau of Economic Research — study on Chivo wallet adoption, 2022 (60%+ of incentive recipients never transacted).
Central American University (UCA) — annual survey on Bitcoin usage in El Salvador, 2023 (88% non-usage) and 2024 (92% non-usage).
El Salvador Chamber of Commerce — survey on business Bitcoin acceptance (86% of businesses never completed a Bitcoin sale).
Chivo wallet remittance figures — $39.4M (2022), $32M (2023), $28.8M (2024) against $2.64B in total remittances over the same span.
International Monetary Fund — El Salvador loan agreement, December 2024 ($1.4 billion; Chivo wind-down and voluntary-acceptance conditions).
El Salvador Legislative Assembly — vote to amend the Bitcoin Law, March 2025 (55–2, private-sector acceptance made voluntary).
President Nayib Bukele — public remarks, December 1, 2024 (“most unpopular measure” acknowledgment).

Stellae Liquiditas LLC · Principal-only digital asset treasury · Stellae Group Division 03. This statement reflects the firm’s own treasury conviction and is not investment advice. Addendum I figures current as of July 17, 2026.