● AI Power Crunch, Nuclear Comeback, Capital Flood
Oct–Dec, the real drivers pushing stock prices are “power, regulation, and capital raising” rather than the bubble
The core points you can confirm directly from my writing are five.
1) The REIT structure, regulatory shortcut tricks, and political risk behind the October listing issue (FRMI).
2) How a Trump-era nuclear restart revalues power stocks.
3) Why Google is buying crypto mining farms, and a field checklist for sites that can realistically pivot to AI data centers and those that cannot.
4) What Oracle’s oversubscribed jumbo corporate bond sale implies about the “health of AI infrastructure capital channels.”
5) The Nvidia–OpenAI ‘infinite-power’ controversy, the decisive difference from vendor financing, and the data points we should monitor.
In short, rather than short-term noise in the US market, I trace in chronological order how AI data centers, nuclear-centered energy transition, and changes in cost of capital will unfold.
1) The essence of the October listing ‘FRMI’ issue: development-stage power and AI infrastructure packaged as a REIT
Start with the IPO selling points.
- It looks abnormal that an AI infrastructure developer less than a year old lists under a REIT shell.
- The strong founder and backer network, a Texas-based large site leased for 99 years, and a design as a ‘power campus’ mixing nuclear, gas, and solar are the core points.
- Early revenue is close to zero, dividends are subordinated, and it is effectively a “developmental REIT” with growth-stock characteristics.
What other YouTube channels and news outlets often miss here is the reason for listing as a REIT.
- For large capital injections into land and power infrastructure during development, REITs can lower cost of capital compared with equity raises.
- However, US REITs are subject to the 75% asset/income tests and a requirement to distribute 90% of taxable income as dividends.
- Therefore, assets with high development risk are often placed into TRS (Taxable REIT Subsidiary) or routed through UPREIT structures to bridge initial cash-flow shortfalls.
- The point is that it is not “stable because it’s a REIT,” but rather a “development project wearing a REIT outfit.”
Claims of four reactors and a 3–5 year commercial operation target require strict fact-checking.
- Historically, obtaining a new large reactor COL (Combined License) from the US NRC often takes 6–10 years.
- ERCOT in Texas has relatively faster interconnection queues, but nuclear is subject to separate federal safety and environmental reviews.
- If the design is a self-contained (islanded) microgrid to reduce grid interconnection dependence, the initial power delivery timeline can be shortened, but you cannot bypass the licensing process for a large nuclear plant.
A quick fact-check checklist to verify promptly.
- SEC S-11/S-1: Whether REIT tests and TRS usage are present, and any conditional clauses in dividend policy.
- NRC licensing path: Whether ESP (Early Site Permit) or COL applications have been filed and the timelines.
- Power mix staged roadmap: Year-by-year MW for Phase 1 gas turbines and batteries, Phase 2 small/large nuclear inputs.
- PPA or fixed-rate contract framework: Customer credit quality, upfront payment structures, and inflation pass-through clauses.
- Cooling water/supply and thermal management (liquid cooling) capacity, and fiber backhaul (communications) availability.
One-line summary.
“FRMI is not dividend-oriented but a ‘developmental infrastructure’ story backed by political and regulatory networks.”
Therefore, valuation should be approached from a growth-stock/project-finance perspective rather than REIT multiples.
2) The market implication of a Trump-era ‘nuclear restart’: the trigger for power stock re-rating
Policy signals are clear.
- Power supply is struggling to keep up with the surge in AI data centers, and policy places nuclear comeback at the forefront.
- Efforts to accelerate many new large reactors and SMR (small modular reactor) pipelines by 2030 are becoming visible.
Who benefits first.
- Merchant generators and utilities with large nuclear proportions gain an advantage in power spreads and PPA price renegotiations.
- Transmission/substation EPCs and extra-high-voltage transformers and switchgear will see longer order books, and delivery lead times may carry a premium.
- SMRs carry significant commercialization timeline risk, but federal support and guarantees (IRA/DOE Loan Programs) increase option value.
Key takeaways to note.
- Power stocks in the US market are being re-rated “like growth stocks.”
- AI data centers are now inseparable from the energy transition.
- Nuclear has increased policy beta as a return to baseload.
3) Google’s ‘mining-farm pivot’ investment: not every site qualifies
Google’s stakes in Terawulf and Cipher Mining are simple to explain.
- Building only proprietary data centers cannot keep up with AI demand.
- Mining farms already equipped with power intake, land, and cooling infrastructure can be rapidly converted to AI hosting (colo/HPC) and brought to market faster.
But not every mining farm can be turned into an AI data center.
Field checklist.
- Power density: 30–60 kW per rack, with some requiring over 100 kW per rack. Retrofitting from air-cooled ASICs requires capex for immersion/liquid cooling.
- Networking: Ability to install Infiniband/400G Ethernet backbone and fiber backhaul.
- Power quality: Handling of momentary outages and voltage sags, redundancy N+1 or better, UPS and battery configurations.
- Regulation & incentives: Local government incentives, participation in demand response (DR) programs, and water regulation.
- Contract structure: Big tech signs long-term deals with prepayments, minimum take requirements, and delay penalties. Does the site have the financial stamina?
The core in one line.
“A pivot succeeds only where the full four-pack of power, cooling, network, and contracts exists.”
If these conditions are unmet, stock prices may jump early but operating results will lag.
4) OpenAI’s ‘computing shortage’ claim and the 125x power demand argument — how the market reads it
A computing capacity shortage is likely to persist over the next 1–2 years.
- Both large-model training and serving are constrained by power, cooling, land, and skilled personnel simultaneously.
- Numbers like “125x” are bold, but the directional signal is clear: power, transmission, and equipment must expand in tandem.
Investment points by chain.
- Generation & power: Players with portfolios of nuclear, gas, and hydro.
- Transmission & EPC: Expanding long-term backlogs, timing for transformers and switchgear additions.
- Power equipment: Switchgear, bus ducts, immersion cooling, chillers.
- Data centers: High-density HPC colos beyond the hyperscalers.
- Semiconductors: Continued accelerator cycles centered on Nvidia, accompanied by networking stack demand.
Lock these five economic keywords in.
The US market’s leadership will be driven by AI data centers, nuclear, energy transition, power infrastructure, and Nvidia.
5) Oracle’s jumbo corporate bond sellout: the ‘funding tap’ is still healthy
A jumbo corporate bond issued to fund data center CAPEX was fully placed despite high interest rates.
- The implication is straightforward: large institutional capital is willing to pay a premium for AI infrastructure bonds.
- If spreads do not widen and issuance is absorbed, other issuers will likely follow.
Watch these warning lights.
- New-issue spreads for high-grade bonds widening sharply by 150 basis points or more.
- Two or more consecutive large deal ‘sell-outs’ failures (deal misses).
If these two occur, the capital-raising channel for the AI infrastructure cycle is starting to wobble.
6) Fact-checking the ‘infinite-power’ controversy: different from vendor financing, but check the indicators
Allegations that Nvidia’s strategic investments and customer sales form a ‘self-sustaining loop’ have circulated, but the essence differs.
- Vendor financing is where suppliers directly extend credit to customers to sustain sales, and that financing flows immediately back into product purchases.
- Current investments in big tech and AI labs are more equity/strategic partnerships and, in most cases, are not directly tied to equipment purchase obligations.
However, the numbers we should check ourselves are these.
- Whether days sales outstanding (DSO) for receivables spikes.
- Expansion of non-standard financing exposure such as customer loans and receivables securitization.
- Overconcentration of revenue by a few large customers and structures of prepayments/rebates.
- Audit note language additions on ‘customer financing (Recourse/Non-recourse)’.
If these indicators deteriorate together, switch to a cautious stance.
7) Chronological investment roadmap: Oct–Dec (0–3 months) → 2026 (12–24 months)
0–3 months: Check earnings, capital markets, and policy signals.
- Earnings: Whether Nvidia, Microsoft, and Oracle maintain AI CAPEX guidance.
- Power: ERCOT reserve margin, wholesale power prices, and transformer lead times.
- Policy: DOE loan guarantees, NRC licensing speed signals.
3–6 months: Reality check on supply-chain expansion.
- Expansion of immersion cooling and switchgear, and utilization of newly opened HPC colos.
- Speed and pricing of long-term PPAs signed by big tech.
6–12 months: Accelerator generation turnover and PPA repricing.
- New accelerator generation adoption increasing per-rack power density.
- Whether inflation pass-through clauses in existing PPAs function as intended.
12–24 months: Transmission permitting bottlenecks and nuclear roadmaps.
- Permitting and construction of major transmission lines, and local NIMBY opposition.
- Milestone achievement rates for SMR and large reactor licensing.
The strategy should be a barbell.
- Defensive axis: Generation and transmission/distribution infrastructure (cash-flow visibility).
- Growth axis: High-density data centers, power equipment, and HPC colocation.
- Selective options: SMR and mining-farm pivots (if conditions are met).
- Risk hedge: Shorten duration against rising rates/spreads and hedge for power-price spikes.
8) Example watchlist and risk notes
Power / utilities (policy beta + cash-flow).
- Constellation Energy (nuclear operating strength).
- Vistra (merchant exposure + IRA credit exposure).
- NextEra Energy (renewables, storage, and transmission portfolio).
Transmission & equipment (pick-and-shovel).
- Quanta Services (EPC backlog).
- Eaton (switchgear / power equipment).
Nuclear / fuel chain (high-risk, high-reward).
- NuScale (SMR; licensing and cost risks are large).
- Centrus Energy (HALEU fuel).
- BWX Technologies (nuclear modules and maintenance).
HPC / data centers.
- Oracle (AI cloud and capital-raising capability).
- Supermicro (servers; cycle volatility is high).
- Mining-farm pivots: Terawulf, Cipher Mining (only if power, cooling, network, and contract requirements are met).
Risk notes.
- Regulatory timeline delays, widening of rates/spreads, local community opposition, cooling/water bottlenecks, and equipment delivery delays.
- “REIT-wrapped development firms” should be valued by project risk and permitting/PPA progress rather than dividend expectations.
9) Quick fact-check guide (when reading new listings and policy news)
- Prioritize original filings: S-1/S-11, 10-Q/K, NRC/DOE documents.
- Break timelines into phases for verification: site → power → cooling → network → PPA → operation.
- “Political branding” is a narrative; cash flows are made by contracts and permits.
- Do not lose sight of two numbers: cost of capital (rates and spreads) and power price.
- The US market is creating real demand via AI data centers and a nuclear-centered energy transition, and the capital markets are still supporting it.
- Developmental REITs like FRMI should be assessed first on permits, PPAs, and capital-raising structure rather than dividend expectations.
- Mining-farm pivots succeed only where the four elements of power, cooling, network, and contracts are present.
- Oracle’s bond sellout signals that funding channels are currently open; convert to caution if spreads widen sharply.
- The ‘infinite-power’ controversy is somewhat overstated; monitor receivables and customer-finance notes directly.
- From Q4 through 2026, a barbell approach with power/transmission as the defensive axis and HPC/equipment as the growth axis is reasonable.
[Related articles…]
- 2025 AI data center investment map: the intersection of power and semiconductors
- Nuclear restart, opportunity points for Korea and the United States
*Source: [ 소수몽키 ]
– 증시 버블론 무시하고 오른다? 트럼프가 대놓고 밀어주는 주식들
● Rich Get Richer, Money Machines, 2025 Economy, AI Roadmap
[Feature Debate Part 3 Commentary] Why Do the Rich Keep Getting Richer? The Structure of Money Making Money, ‘How to Stay Rich’, and the 2025 Global Economy & AI Trend Roadmap
This article contains four points that are rarely covered on YouTube or in the news.
First, how to read polarization by “wealth replacement rate” rather than “fixed rich” and a proposal for a Korean-style indicator design.
Second, a structural map of the “power of capital” that dissects the core engines of money making money into financial leverage, code (software/AI) leverage, and media leverage.
Third, how to design a personal operating system (OS) for “staying rich” centered on cash-flow flows rather than asset stock.
Fourth, a timeline of the global economy outlook and AI trends from 2020 to 2026 and the actual signals they send to asset markets.
1) Constants and Variables of Capitalism: Why Polarization Arises and What to Watch
Constant: As markets develop, information spreads faster, and even a slightly better quality or experience produces a “winner-takes-most (more precisely winner-most)” outcome.
It’s the same reason Apple maintains high margins for extended periods.
For this reason, polarization and the concentration of wealth tend to increase structurally.
Variable: The important thing is not “who is rich” but “how often the rich are replaced”.
If the same top 1% are the same people for 5 or 10 years, that’s stagnation; if replacement is active, that’s dynamism.
Korean-style indicator proposal: 5-year/10-year retention rates for the top 1% and top 10% cohorts, income-bracket mobility matrices, startup→EXIT→re-entry turnover rates, intergenerational asset transfer speed, and distribution of ages for first home and stock purchases.
If these indicators improve, polarization may remain but the ladder works.
Key perspective: Policies that enhance “ladder effectiveness” are materially important rather than just changing the “rich ratio”.
2) Global Economy Outlook from 2020 to 2026 and Asset Market Signals by Timeline
2020–2021: Pandemic stimulus caused excess liquidity and inflation surged due to real-economy constraints.
Asset prices rose together, and it was a classic period where “money made money”.
2022–2023: Interest rate hikes occurred simultaneously around the world.
Valuation rebalancing and sharp increases in debt costs led to a regime of risk-selective markets.
2024: Regional policy asymmetry intensified due to differences in the pace of disinflation.
The United States showed resilient consumption/productivity, Europe showed slowdown and energy sensitivity, and China experienced restructuring pains.
2025–2026 scenarios: Disinflation is the baseline, but wage and service-price stickiness remain.
Gradual rate cuts are the base case, but the shadow of “higher rates for longer” persists.
AI trends will create an actual productivity inflection point, widening profit-cycle divergence.
Asset market point: The core of the global outlook is not a simple bet on interest-rate cuts but the “quality of cash-flow recovery”.
The real estate market hinges less on rate-cut expectations and more on rental demand/vacancy/Cap Rate rebalancing.
3) The Engines That Make Money Make Money: Dissecting the Power of Capital into Three Leverages
Financial leverage: In ranges where r>g (rate of return on capital > growth rate), leverage steepens the slope of compounding.
Tax deferral, compounding period length, and default management are crucial.
Code leverage: Software and AI have marginal costs that approach zero.
By using code to replicate time, individuals can mimic the “returns structure of capital owners”.
Media leverage: Reach expansion based on trust reduces customer acquisition cost (CAC) and raises capital efficiency.
Brand is long-term capital.
Conclusion: Modern capital is not just money but also code and trust.
Combining at least two of the three leverages allows one to catch up to the speed at which “money makes money”.
4) How to Become Rich vs How to Stay Rich: Designing a Flow-Centric Personal OS Instead of Focusing on Stock
Redefinition: Wealth is freedom from anxiety rather than the absolute amount.
The core is not “you can play from now” but “money is not the number one stressor in life”.
Flow perspective: Obsessing over stock (total assets) leads to simultaneous increases in consumption and expectations, causing anxiety to recur.
Decisions should be based on cash flow and margin of safety to “stay rich”.
Personal OS design: A buffer of three years’ living expenses in cash/MMF, automatic caps on variable spending, three income streams (main job, capital income, code/content), a fixed “stable cash cow” slot in the portfolio, and use leverage only within ranges coverable by cash-flow schedules.
5) 12-Week Practical Checklist: A Playbook That Fits Reality
Weeks 1–2: Prepare a personal profit and loss statement and cash-flow statement.
Tag monthly fixed costs, variable costs, and capital expenditures separately.
Weeks 3–4: Separate a safety-buffer account.
Physically separate a paycheck account from investment and buffer accounts.
Weeks 5–6: Build a “layered portfolio”.
Partition into cash, bonds, dividend/rental cash cows, growth, and experiment (≤5%).
Weeks 7–8: Document leverage guidelines.
Define your own maximum LTV, DSR, and interest-rate stress-test numbers.
Weeks 9–10: Introduce code leverage.
Automate repetitive tasks with AI agents, RPA, or scripts, and record results as SOPs.
Weeks 11–12: Start media leverage.
Publish one article/video/newsletter per week to build “trust capital”.
6) Policy and Social Design: Public Choices to Increase the “Wealth Replacement Rate”
Fair entry: Lower the costs of starting a business and re-entry, and simplify bankruptcy and recovery systems.
Capital markets: Expand access to initial capital through small public offerings, crowdfunding, and tokenized securities.
Tax policy: Strengthen tax credits for labor and entrepreneurial risk, provide incentives for long-term holdings and investments in real productivity, and recycle excess unearned income through taxation.
Mobility metrics: Regularize top-1% retention rates and income-bracket mobility rates as core national reports.
Make sure citizens can see whether those numbers improve year by year.
7) Real Estate vs Stocks vs AI Capital: Where and How Now?
Real estate market: Don’t focus only on rate-cut expectations; check whether NOI (net operating income), vacancy, and Cap Rate rebalancing are complete.
For residential, look at regional population and supply gaps; for income properties, cash flow and rent-refixing clauses are key.
Stocks: Core holdings should be sectors with high earnings sensitivity and productivity leverage (cloud, semiconductors, software, industrial automation), and use dividend/dividend-growth stocks to manage volatility.
AI trends: Companies that embed AI into workflows to reduce actual time spent will be revalued for margins rather than the model itself.
Individuals should also automate tasks with AI agents and convert “time surplus” into cash flow.
Principle: Avoid overconcentration in the same sector or the same factor, and predefine rebalancing rules for regime shifts (rates/dollar/policy).
8) Common Misconceptions Corrected: The Traps of Anxiety and Comparison
“If you have 10 billion won, anxiety disappears”: Consumption standards automatically rise with income and assets.
Anxiety is reduced not by an absolute amount but by structure (buffer, cash flow, leverage calendar).
“The rich don’t share secrets”: The core is not secrets but systems.
Documenting cash flow, risk, automation, and taxes is competitive advantage.
“Real estate and stocks are completely different”: Both are priced by liquidity, interest rates, and discounted rents/earnings.
The difference lies in operating intensity and the degree of information asymmetry.
9) Key Takeaways from the Debate: Recognize Lawful, Moral Wealth and Spread ‘Possibility’
Sentiment: Lawfully and morally accumulated wealth should be socially recognized.
Only then will reinvestment, job creation, and tax revenue circulate in a virtuous cycle.
Psychology: The belief “I can do it too” is the starting point of wealth.
The more people who learn and take risks, the higher the overall productivity of society.
Practice: Shift consumption and investment grammar toward staying rich rather than merely becoming rich.
10) Six Actions for Today
1) Write down a target buffer equal to 36 months of living expenses and create a dedicated account.
2) Draw a roadmap for three income streams (main job, capital, code) and set quarterly goals.
3) Fix a “cash-flow slot” in your portfolio and increase dividends, rents, and coupons.
4) Run leverage stress tests under three interest-rate scenarios and document them.
5) Automate three work tasks with AI and reallocate the saved time to income-generating activities.
6) Once a year, evaluate your personal “wealth replacement rate” metrics.
(Rate of capability expansion, diversification of income sources, growth of trust capital, recovery speed after failure).
11) Keyword Check: What to Include Today from an SEO Perspective
Global economic outlook, inflation, interest rate hikes, real estate market, AI trends.
These five keywords run through every axis of today’s content.
12) Conclusion: Widen the Ladder and Upgrade Your OS
Polarization is a constant, but replacement rate is a variable.
Policy should widen ladders, and individuals should upgrade their operating systems.
Combine the leverages of money, code, and trust, and design around “flow not anxiety”.
That is the most realistic way to stay rich.
< Summary >
Polarization is structural, but the real issue is the “wealth replacement rate”.
The engines that make money make money are the combination of financial, code, and media leverage.
You must eliminate anxiety with a flow-centric personal OS rather than focusing on stock to “stay rich”.
2025–2026 will be a period of rate easing and AI-driven productivity divergence, so the quality of cash flow will be the battleground.
Policy should widen ladders, and individuals should document and execute on leverage and risk.
[Related posts…]
2025 Global Economic Outlook and Dollar Cycle Transition Signals
The Datafication Strategy of the Real Estate Market Transformed by AI Trends
*Source: [ 경제 읽어주는 남자(김광석TV) ]
– [특집 토론 3편] “왜 우리는 부자가 되지 못할까?” 사람들이 놓친 돈이 돈을 버는 구조와 자본의 힘 | 이광수 대표, 이진우 기자
● Bitkey Redefines 2025 Bitcoin Self Custody, Seedless Security Unleashed
“Never hacked?” Security in the real world is a probability game: The essence and opportunity of 2025 Bitcoin self-custody redefined by Bitkey
This article covers multi-sig designs that recover without a 12-word seed, why inheritance features become a benchmark for judging a “primary vault,” regulatory and censorship risks introduced by server keys and response strategies, defenses against social engineering in the AI era, and the correlation between the 2025 macro environment (interest rates, inflation, liquidity) and hardware wallet demand.
It also unpacks the gray zones of “recovery governance” and an operational security checklist in practical terms — topics other news outlets and YouTube often miss.
The interviewee is Jonathan Pollack, Bitkey product lead at Block, who also disclosed data showing Korea is one of the top purchasing countries.
1) 2024–2025 background: a threefold complex environment of ‘macro + regulation + AI threats’
Bitcoin price volatility is a function of liquidity cycles, the interest rate path, and inflation expectations.
2025 is a period when risk-asset preferences are being reorganized alongside discussions of a US rate pivot, and this will determine the speed of Bitcoin and blockchain adoption.
As ETF inflows and on-chain activity increase, “where to store my assets” becomes a core variable in investment strategy.
Meanwhile, AI-enhanced phishing, deepfakes, and SIM swap social engineering attacks are exploding, increasing the attack surface of software wallets.
This combination structurally pushes up demand for hardware wallets.
2) Why self-custody: the core is “design of responsibility” rather than exchange risk
The intrinsic strength of Bitcoin is that anyone can store and transfer assets themselves on public infrastructure.
Keeping assets on an exchange is convenient, but it exposes you to external risks like credit issues, hacks, withdrawal limits, legal seizure, and financial soundness.
Self-custody is heavy with responsibility, but conversely it maximizes control, censorship resistance, and the intensity of private ownership.
3) Bitkey structure: 2-of-3 multisig and recovery without a seed
Bitkey adopts a 2-of-3 multisig where two of three keys are required to sign a transaction.
Key 1 is a hardware device, Key 2 is a smartphone, and Key 3 is Block’s server recovery key.
Under normal conditions, transactions can be processed independently with “my two keys (hardware + phone),” and Block cannot unilaterally authorize transactions.
If a device is lost, the remaining personal key plus Block’s recovery key can migrate funds to a new wallet.
The key point is that Bitkey does not entrust the 12-word seed (mnemonic) to customers, and this differentiation has driven market enthusiasm.
4) Fundamental difference from other hardware wallets: redesigning the “backup UX”
Traditional hardware wallets require customers to store a 12/24-word seed on paper or metal plates.
The problem is that users must shoulder all real-world risks — fire, flooding, theft, exposure, family inheritance — making actual operational difficulty high.
Bitkey introduces a server recovery key based on multisig so backup responsibility is not placed entirely on the individual.
This design lowers the entry barrier and significantly reduces the psychological hurdle to moving assets into a “primary vault.”
5) An important but underdiscussed topic: ‘recovery governance’ and regulatory risk
The server recovery key is convenient, but in certain jurisdictions, under sanctions, or by court order, service cooperation may be restricted.
Under normal circumstances you can autonomously send funds with your two keys, so the ongoing risk of censorship is low, but there is a scenario where recovery could be blocked when needed, and that must be managed.
Countermeasures include geographic diversification of key distribution, separation of co-signers, holding backup devices, and setting self-discipline routines that maintain two keys without needing recovery.
This point is often missing from marketing, but real security is completed by governance rather than by features alone.
6) Trade-offs of screenless hardware: security UX decides success
If a device lacks its own display, transaction information must be verified through the phone UI.
When a phone is compromised, the danger that “what you see” and “what you sign” differ increases, so the workflow must be designed to mitigate this.
Practical tips include enforcing amount limits, delayed transfers, and multi-step confirmation routines for large transfers, and operating a separate wallet for small amounts.
These operational security habits create perceived security beyond hardware performance.
7) Demand trends: a function of price, inflows, and risk awareness
Jonathan Pollack said that the message of “recovery without a seed” and ease of use are the main demand drivers.
As Bitcoin prices rise, sensitivity to custody risk increases, and security stages naturally escalate from exchanges → software wallets → hardware wallets → multisig.
Korea and the Asia-Pacific region were mentioned as top-selling regions, reflecting sophisticated retail investors and rapid learning curves.
8) The significance of inheritance features: a measure of whether it is a “real primary vault”
Bitkey’s inheritance feature is a strong signal showing whether customers use the wallet as a generational asset store rather than a day-to-day spending wallet.
Bitcoin inheritance is legally and operationally complex, involving jurisdiction, personal data, and key disclosure risks, and a productized inheritance flow greatly lowers the user’s psychological barriers.
Operational tips are to periodically review heir identification information and procedures, and to update legal documents and on-chain access rights together.
9) 2025 macroeconomy and the correlation with hardware wallet demand
The greater the expectation of an interest rate decline, the more liquidity is reallocated, and when risk-asset preference recovers, trading volume in crypto and custody demand increase together.
If inflation expectations rise again, the “digital gold” narrative strengthens and the emphasis on distributed storage of Bitcoin grows.
When US equities and digital assets move under the same risk premium, portfolio rebalancing will generate new inflows to hardware wallets.
10) Future revenue models: shifting from ‘custody’ to ‘utility’
Bitkey’s current revenue is concentrated on hardware sales, but over the next 10–15 years “financial services built on Bitcoin” will create added value.
Layers such as lending, insurance, derivatives, and claim management implemented on top of self-custody will open new fee-based economic zones.
The key is designing services so users can utilize them “without surrendering their keys,” and that will be a real turning point for web3-style finance.
11) Scalability and fees: between Layer 2 and fee cycles
Block-affiliated Spiral and others continue researching Layer 2s like Lightning, which reduce perceived transfer costs.
The interview also mentioned “recently low fees,” but fees cycle with on-chain demand, hash price, and block space competition.
In the long run, integration of L2 infrastructure and wallet UX will determine mainstream usability.
12) Privacy roadmap: structures that make balances and activity invisible
The Bitkey team presented privacy enhancement as a roadmap pillar so that even if they provide the app and recovery key, they cannot see customer balances or transaction history.
This includes minimal data design between wallet and server, metadata reduction, geographic dispersion, and regulatory responses.
Practical tips are to avoid address reuse, consider lawful use of privacy tools like CoinJoin, and separate deposit and withdrawal paths.
13) Operational security checklist for the AI threat era
Assume deepfake voice or video requests for urgent transfers and adopt time delays and two-person approval policies for large transfers.
Buy hardware wallets only from authorized distribution channels and record unboxing and firmware verification procedures.
Perform NFC pairing only in fixed locations and keep devices separated when not in use.
Use a phone dedicated to financial activities, prohibit rooting/jailbreaking, use biometric plus PIN dual layers, and replace SMS 2FA with FIDO security keys.
14) Risk registry: know it to avoid it
Legal and regulatory: identity verification requirements during recovery and the possibility of service restrictions in certain jurisdictions.
Supply chain: the risk of opening or tampering during delivery, so be sure to establish integrity verification routines.
Physical threats: set transaction amount limits, delay locks, and distribute multisig locations to prepare for coercion scenarios.
Phone malware: add human procedures such as reading the recipient address and amount aloud to verify before approval.
15) Practical setup suggestions for Korean investors
Wallet separation: dualize into a daily-use wallet (small amounts, hot) and a vault wallet (Bitkey, multisig).
Limit policies: vault wallets should have per-transaction and daily limits, delays, and allow only whitelisted addresses.
Inheritance documentation: configure inheritance features and reflect procedures and contact information in legal documents, reviewing quarterly.
Macro alignment: adjust custody strategies around interest rate events and minimize exchange exposure time.
16) Summary conclusion: “There is no absolute safety, but structure and habits change probabilities”
Bitkey’s 2-of-3 multisig and recovery without a seed have changed the paradigm of “backup UX.”
The server recovery key provides convenience, and the dual personal keys preserve sovereignty.
The decisive factor is institutionalizing recovery governance risk and operational security, which determines qualification as a primary vault.
Given 2025’s interest rate and inflation conditions, hardware wallet demand is likely to increase structurally, and Korea is at the forefront of that trend.
In the AI era, upgrade both technology and habits together.
Timeline summary: header — details — key points
1) The 2024–2025 macro environment emerges.
– Interest rates, inflation, and liquidity changes expand Bitcoin custody demand.
– AI-based attacks increase software wallet risks.
2) Bitkey background and structure.
– 2-of-3 multisig, seedless recovery, Korea as a top purchasing country.
3) Differences from other wallets.
– Structurally reduce customer backup responsibility and include inheritance features.
4) The unseen core.
– Managing recovery governance and regulatory risk is the key to long-term security.
5) Operational security and AI response.
– Delays, limits, two-person approvals, device separation, supply chain verification.
6) Connecting to investment context.
– Optimize custody strategy around US market and Bitcoin dynamics and interest rate events.
< Summary >
Bitkey redesigns the “backup UX” with 2-of-3 multisig and seedless recovery to lower the barrier to self-custody adoption.
The core differentiators are inheritance features and recovery governance, which encourage large-scale transfers into a primary vault.
However, the regulatory and censorship risks of a server recovery key must be managed as scenarios.
Under 2025’s interest rate and inflation environment and with AI threats, hardware wallet demand is likely to rise structurally.
The conclusion is simple.
When technology (multisig) and habits (operational security) combine, the probability of being hacked is greatly reduced, even though nothing is absolutely safe.
[Related articles…]
2025 Bitcoin and the interest-rate cycle: ETF inflows and the dynamics of liquidity
AI and blockchain security: asset protection strategies in the deepfake era
*Source: [ Jun’s economy lab ]
– 절대 해킹될 수 없는 비트코인 지갑입니다(ft.Jonathan Pollack of BLOCK Product for Bitkey)
● Shutdown Shock, Q4 Rally Setup, AI Power Crunch, Robinhood Prediction Boom, MLB Price Squeeze
Early October US shutdown risk, Q4 market seasonality, the real bottlenecks of AI capex, Robinhood’s “prediction market” revenue model, and MLB game attendance prices: complete summary of exclusive key points in this piece
This article includes where markets would actually be disrupted if a shutdown occurs on October 1 (Treasury issuance, data gaps, Fed decision delays), the historical Q4 seasonality and contrarian buying windows, the hidden bottlenecks that will sustain the AI investment cycle through mid‑2026 (power, HBM, packaging, cooling), the regulatory and revenue structure behind Robinhood’s prediction market that has lifted its stock since S&P 500 inclusion, and why MLB in‑person attendance has become a “special event” for the American middle class.In particular, it highlights “real variables” that are less often discussed elsewhere: the Treasury issuance calendar and short‑term rate microstructure (repo, T‑bill), substitute indicators that matter during official data gaps, CFTC risk for prediction markets, and power/cooling bottlenecks for AI data centers.It is organized to the end in chronological order from global economy, rates, inflation, US equities, to the dollar.
9/30~10/1: Shutdown imminent checklist and immediate variables
-
Header: After 9/30 JOLTS and Conference Board consumer confidence releases, if no budget agreement by 0:00 on 10/1 a federal government shutdown may begin.
-
Details: impacts in the event of a shutdown.
-
Main takeaway: data gaps and changes in bond market microstructure matter more than an immediate “index crash.”
-
Government statistics gap risk.If most federal statistics are delayed, the October 3 Nonfarm Payrolls (NFP) release could be postponed.If CPI, retail sales, etc. are delayed, the Fed will rely less on hard data and put more weight on alternative indicators like card spending, payrolls, and job posting data.If a backlog of data is released in the first week after data resumes, volatility (VIX) could spike belatedly.
-
Bond and dollar microstructure.Even if fiscal spending halts, Treasury interest payments continue.The problem is the Treasury’s short‑term funding operations.If the timing of CMB (additional short‑term Treasury issuance) is disrupted, T‑bill supply/demand can become rigid and GC repo rates may spike with a temporary “liquidity premium.”The dollar index faces a tug‑of‑war between “risk‑off strong dollar” and “growth‑slowing weak dollar,” but in practice the former is likely to dominate initially.
-
Sector sensitivity (short term, 1–2 weeks).R&D/consulting/IT firms with high federal contract exposure, some data processors, and loan channels tied to credit guarantees will suffer delays.By contrast, defensives (utilities, consumer staples) and high‑dividend telecoms will act as stabilizers.With indicator gaps, “AI‑beneficiary large caps” may retain story‑driven buying, but without guidance their volatility can increase.
10/3~10/10: Investment operations when employment and inflation releases are delayed
-
Header: operating routine during data delay.
-
Details: substitute indicators, positioning, volatility.
-
Main takeaway: in a “market without data,” price action and alternative indicators gain influence.
-
Priority alternative indicators.Monitor ADP payrolls, Homebase hours worked, Indeed job postings, card issuer spending, container logistics, and airline/hotel fares as high‑frequency inflation signals.Regional Fed business surveys (Philly/Empire) and the ISM price index acceleration are critical.
-
Position management.For options, instead of buying short‑dated calls/puts outright, using calendar spreads and butterflies timed to the week of data resumption is a more cost‑efficient way to price jump risk.For equities, maintain a core (index + quality) plus satellite (themes) structure, reduce satellite weights, and lower leverage — a conservative stance is advised.
October~December: historical seasonality of Q4 and contrarian entry windows
-
Header: historical patterns of Q4 performance and practical interpretation.
-
Details: October volatility, November–December rally potential.
-
Main takeaway: October declines or sideways moves often become the “entry cost” for November–December gains.
-
Fact‑check points.Historically, average returns in Q4, especially in November and December, have been favorable, and October shows strong recovery resilience relative to volatility.This year, shutdowns, data gaps, and conservative earnings guidance could increase October volatility, but liquidity, year‑end rebalancing, and resumed buybacks could support the downside.
-
Strategic allocation.Weeks 1–2 of October: keep event hedges, increase defensive and quality exposure, hold cash‑like assets.Mid‑to‑late October: after reading the tone of earnings guidance, re‑expand exposure into a potential November–December rally (large tech with relative strength, industrials, and semiconductor supply chain).
The real bottlenecks that will sustain the current AI rally through mid‑2026
-
Header: the cycle is determined by quantity and infrastructure rather than price.
-
Details: power, HBM, packaging, optics, cooling.
-
Main takeaway: the reason AI capex does not show signs of slowing is supply constraints, not demand.
-
Power / substation capacity.For large data centers, grid power availability is a bigger bottleneck than available land.Substation and transmission upgrades, PPA (power purchase agreement) execution, and delays in switchgear and transformers determine 18–30 months from groundbreaking to rack activation.Demand for utility and transmission equipment and for SiC/GaN power semiconductors is structurally rising.
-
Memory / packaging.HBM3E buildout and CoWoS/FC‑BGA packaging capacity are tight.Even if wafer production increases, packaging equipment and skilled packaging engineers cannot be rapidly scaled.This tends to lock in GPU and AI accelerator preorders and long‑term contracts into 2025–2026 and supports ASPs.
-
Networking / optical modules.Migration to 800G optical transceivers and new switch chips is a bottleneck.In data center east‑west and north‑south topologies, shortages can cascade down to cables and connectors.
-
Cooling standardization.Shift from air cooling to direct liquid immersion, radiators, and CDU adoption is accelerating.Immersion can raise per‑rack power density 2–3x, but maintenance and standardization issues will persist, so a mixed deployment through 2026 is likely.
-
Key sentence for investors.The AI cycle is driven long term by the speed of quantity and infrastructure expansion, not by price narratives.Even if global macro conditions gradually ease rates and inflation, AI infrastructure constraints will operate independently.
Why Robinhood’s prediction market lifted the stock: revenue model, regulation, and sustainability
-
Header: the essence of the new business that sustained momentum after S&P 500 inclusion.
-
Details: usability, fees, risks, regulation.
-
Main takeaway: prediction markets are engagement products that increase LTV by extending time on platform.
-
Product mechanics.Users bet yes/no on real‑world events (elections, sports, FOMC, inflation prints) under a contract structure.A small fee per contract (for example, a few cents) is shared between the platform and operator, and as volumes rise, margins improve versus fixed costs.The daily stickiness that makes users return to check issues is the core value.
-
Risks and differentiators (important).Regulation: contracts overlap with CFTC rules on event contracts.Whether political and economic event contracts are classified as “gambling” or as “derivatives” will be a recurring debate.Some platforms have previously suspended or restricted political markets, and product sets can change quickly based on regulatory interpretation.Market making and liquidity: if prices cluster at 50:50 and slippage is large, informed participants will capture profits and casual users may churn.Tax/reporting: users may accumulate frustration over tax handling of many small trades.
-
Opportunity.There is a new demand for “information hedges.”For example, if you cannot sufficiently reduce portfolio delta ahead of a rate cut/hold scenario, event contracts can provide partial hedges.If institutional advisors and retail alike become comfortable using them for this purpose, structural volumes could follow.
Why MLB in‑person attendance has become a “special event” for the American middle class
-
Header: it’s about revenue model changes, not ticket price alone.
-
Details: FCI, dynamic pricing, RSN collapse, on‑site revenue diversification.
-
Main takeaway: team revenue models have shifted toward maximizing on‑site ARPU.
-
The reality of price structure.Average tickets for popular teams like the Yankees and Red Sox commonly exceed about $150 per person.Parking runs $40–$80, and F&B and merchandise prices are rigidly marked up via dynamic pricing.This is why the Team Marketing Report’s Family Cost Index (FCI) sets new highs every year.
-
RSN (regional sports network) collapse spillovers.With cable bundle breakups and some RSNs failing (making broadcasting rights revenue uncertain), teams have strengthened direct monetization through in‑venue and streaming sales.As a result, on‑site ARPU and the share of premium seating have increased, and due to supply and fee structures general admission prices are felt to be higher.
-
Contrast with Korea (per income feel).In Korea, mass transit access, low‑priced outfield seats, and fan culture make attending more of an everyday extension.In the U.S., attendance is positioned as a “celebration or gift,” so spending relative to median income has risen.Even if global economic conditions cool, in‑person demand is sustained by concentrated premium demand, and prices are unlikely to fall easily.
Q4 practical checklist at a glance
- Rates / dollar: early in a shutdown and data gap, an initial dollar strength and mixed short‑ vs. long‑term rates is the base scenario.
- Inflation: energy, rents, and service price slowing is the medium‑term baseline, but pay attention to inflation expectations if wage data are delayed.
- US equities: October volatility can be an opportunity.Liquidity and rebalancing are likely to be supportive into November–December.
- AI: watch quantity bottlenecks rather than price.Power, HBM, packaging, optics, and cooling are the checkpoints that determine equity staying power.
- Themes: prediction markets can be a killer feature that raises Robinhood’s dwell time and ARPU, but always price in CFTC regulatory headline risk.
< Summary >
- If a shutdown occurs on 10/1, the initial impact will show up as a “data gap” and “short‑term liquidity” stress.Watch T‑bills, repo, and the dollar first.
- October is volatile but statistically often serves as an entry window for November–December gains.
- The AI rally is sustained by infrastructure bottlenecks rather than price.Monitor power, HBM, packaging, optics, and cooling as checkpoints.
- Robinhood’s prediction market is a fee‑based, high‑margin, high‑stickiness model, but CFTC regulation and liquidity quality will determine success.
- Rising prices for attending MLB games are the result of teams maximizing on‑site ARPU after the RSN collapse.For the middle class, games have become a “special event.”
[Related posts…]
If the U.S. government shutdown is prolonged: implications for the dollar, rates, and equities
Prediction market innovation: how Robinhood Prediction Market is changing investor habits
*Source: [ Maeil Business Newspaper ]
– 합의못하면 美정부 10월1일 셧다운ㅣ로빈후드 ‘예측시장’ 주가 견인ㅣ미국 중산층도 가기 힘든 MLB직관ㅣ홍키자의 매일뉴욕