Global Awareness, Market Alignment, and the Anatomy of Failure

Abstract

This research paper provides an exhaustive investigation into the structural, geopolitical, and economic causes behind the collapse of technology, software, and Artificial Intelligence (AI) startups between 2020 and 2025. Moving beyond the reductive statistic that "90-99% of startups fail," this report leverages the proprietary framework of the Market Egregore: the collective, emergent intelligence of the market system; to diagnose why well-funded, highly innovated ventures succumb to extinction. Through a rigorous analysis of thirty specific failure cases, including high-profile collapses such as Northvolt, Graphcore, Cerebras Systems (IPO withdrawal), and Nook, the paper establishes that the primary drivers of mortality were not product deficiencies, but a catastrophic lack of Global Awareness and Market Alignment. The research dissects the impact of specific 2025 vectors: the punitive H-1B visa policy shifts, the "October 2025" mini-recession and government shutdown, and the unhedged exposure to volatile energy markets. By synthesizing quantitative failure data with qualitative storytelling and theoretical modeling, this report offers a new paradigm for understanding startup survival: success is not merely a function of innovation, but of synchronization with the geopolitical and macroeconomic "consciousness" of the global market. {Check out the Beunec Platform}

Definitions

Market Egregore:

Defined in the internal publication The Order of Market Egregore (Akinnawo, 2025), the Market Egregore is the market viewed not as a mechanism, but as a living, evolving entity. It is a collective thought-form powered by the belief, trust, fear, and capital of humanity. It operates on a prime directive: "The Market Always Wins." A market crash or recession is not a failure of the system, but a biological "purge" or self-correction, wherein the Egregore sheds inefficient capital allocations to reset the conditions for future growth.

Market Alignment:

The strategic synchronization of a venture’s operational cadence with the external rhythms of the Market Egregore. Unlike Product-Market Fit (PMF), which focuses on the user-product relationship, Market Alignment focuses on the venture-environment relationship. It encompasses adaptability to interest rate cycles, geopolitical friction, regulatory tokenization trends, and energy availability.

Unit Economics (The 2025 Context):

The direct financial relationship between the cost of delivery and the value captured. In the 2023-2025 period, this definition expanded to include the "embedded costs" of geopolitical risk; such as the rising cost of compute due to energy volatility and the rising cost of talent due to immigration barriers. A startup with positive margins on paper but unhedged exposure to these variables is considered to have "Broken Unit Economics".

Absolute ROI:

A shift in Venture Capital psychology observed in 2025, moving away from "Relative ROI" (growth compared to competitors) to "Absolute ROI" (cash-on-cash returns adjusted for risk-free rates). This shift precipitated the "Capital Strike" that starved late-stage ventures of liquidity.

Introduction: The Silent Cull of 2025

The technology sector entered 2025 with the optimism of a bull market, fueled by the generative AI boom of the preceding two years. Yet, by November 1, 2025, the landscape was littered with the wreckage of "unicorns" and "decacorns" that had seemed invincible only months prior. This was not a random distribution of failure. It was a systematic culling, a targeted removal of organisms that were maladapted to a suddenly hostile environment.

Conventional wisdom, codified in business schools and accelerator programs for decades, posits that startups fail primarily because they "build things nobody wants" or "run out of money." While statistically true, these explanations are symptomatic, not causal. Why do they run out of money? Why does the market demand evaporate? Our analysis of the 2023-2025 failure cohort suggests that the root cause lies in a fundamental misunderstanding of the Market Egregore.

Founders and investors operated under the assumption that the market was a passive environment, a playing field upon which they could impose their will through sheer force of capital and innovation. They failed to recognize the market as an active, predatory intelligence that reacts to imbalances. When billions of dollars flowed into "AI wrappers" like Nook or Aura Labs , the Egregore did not passively accept this liquidity; it metabolized it, extracted the useful data, and purged the host companies. When Northvolt attempted to scale battery manufacturing in a high-cost European energy environment , the Egregore did not support the political narrative of "green sovereignty"; it crushed the inefficiency with the brutal logic of unit economics.

This report is structured to deconstruct these failures layer by layer. We begin by establishing the theoretical framework of the Market Egregore. We then proceed to a forensic analysis of the specific geopolitical and economic executioners of 2025: the Immigration Shock, the Energy Crisis, and the Capital Strike. Finally, we present detailed case studies of the most significant failures of the era, weaving together the quantitative data of their collapse with the qualitative narrative of their strategic blindness.

Dialogue: The Boardroom Reality Check

Setting: The sleek, mahogany-paneled boardroom of a Tier-1 Venture Capital firm in Menlo Park. It is late October 2025. Outside, the sky is grey, mirroring the economic sentiment. Inside, a Founder is pleading for a bridge round.

Founder (Series C, AI Infrastructure): "You can't pull the term sheet now. We have the H100s. We have the team. Our model benchmarks are within 5% of GPT-5. We just need six more months to hit profitability."

Partner (VC): "It’s not about your benchmarks, David. Look at the CPI prints. Look at the bond yields. The government shutdown has frozen three of your five biggest government contracts. The new visa fees have blown a 15% hole in your OpEx because half your engineers are on H-1Bs. You aren't fighting us; you're fighting the gravity of the situation."

Founder: "But the technology... the innovation... the market needs this!"

Partner: "The market doesn't need anything. The market is an entity that optimizes for survival. Right now, survival means liquidity and risk-off assets. You are a risk-on asset in a risk-off world. The Egregore has shifted, and you didn't shift with it. We're not writing the check."

This dialogue encapsulates the central thesis of this report: Innovation without alignment is merely an expensive hobby.

Section 1: The Order of Market Egregore

To understand why so many startups failed in this specific window, we must first understand the nature of the system they were operating within. The "Market Egregore" is a theoretical framework developed at Beunec Technologies to describe the non-linear, emergent behavior of the global economy.

1.1 The Anatomy of the Beast

The market is often described mechanically: supply, demand, equilibrium. However, mechanics implies predictability. The market of 2025 proved definitively that it is biological and psychological. It is an Egregore, a psychic entity created by the collective focus of a group. In this case, the group is the entire human race participating in economic exchange.

The Egregore functions on a cycle of intake, digestion, and purge.

• Intake (The Boom): The Egregore entices participation through the release of "dopamine": in the form of easy capital, hype, and valuation bubbles. The AI boom of 2023-2024 was a classic intake phase. Money flowed into Graphcore, Cerebras, and thousands of Y-Combinator startups.

• Digestion (The Test): The market tests these inputs against reality. Can Nook actually reduce sales cycles, or is it just an annoying bot? Can Northvolt actually produce batteries cheaper than CATL?

• Purge (The Crash): The "October 2025" mini-recession was the purge. The Egregore identified the inputs that failed the digestion test and expelled them. The bankruptcy of Stenn (Fintech) and EasyMile (Mobility) was the system removing toxins.

1.2 The Myth of "Market Failure"

When a startup fails, we call it a "market failure." This is a misnomer. From the perspective of the Egregore, the failure of a startup is a "Market Success."

Consider the collapse of Graphcore. The company raised over $700 million to build a competitor to Nvidia. It failed and was acquired for pennies. To the founders and investors, this was a tragedy. To the Egregore, this was a victory. The market determined that the most efficient allocation of resources was to consolidate around the Nvidia/CUDA ecosystem rather than fragmenting into a second architecture. By bankrupting Graphcore, the Egregore freed up the talent (engineers) and capital (remaining assets) to be redeployed to more productive areas. The "Order" was maintained.

1.3 The Hierarchy of Alignment

For a startup to survive the Egregore, it must achieve alignment at four distinct levels. The failures of 2025 can be categorized by the level at which they failed.

• Level 1: Product Alignment. (Does the user want it?)

  • Failure Example: LumenOrbit. Lacked a clear product-market fit.

• Level 2: Economic Alignment. (Do the unit economics work?)

  • Failure Example: Nook. High user acquisition costs (CAC) exceeded Lifetime Value (LTV).

• Level 3: Ecosystem Alignment. (Does it fit the technological zeitgeist?)

  • Failure Example: Graphcore. Built great hardware that was incompatible with the dominant software (CUDA).

• Level 4: Geopolitical Alignment. (Does it fit the macro-regulatory reality?)

  • Failure Example: Northvolt. Failed due to energy prices, tariffs, and supply chain wars; factors entirely outside the product itself.

The 2023-2025 period was unique because the mortality rate at Level 4 (Geopolitical) skyrocketed. Founders who were geniuses at Level 1 (Product) were slaughtered at Level 4 because they lacked Global Awareness.

Section 2: The Geopolitical Executioners of 2025

The years 2023 through 2025 saw a convergence of hostile macroeconomic vectors that targeted the specific vulnerabilities of early-stage technology companies. Unlike previous downturns which were often sector-specific (e.g., the Dot-com crash), the 2025 contraction was structural, driven by policy, energy, and labor shocks.

2.1 The Immigration Shock: The H-1B Visa Guillotine

In 2025, the regulatory environment regarding high-skilled immigration in the United States underwent a radical transformation. For decades, the H-1B visa program had been the lifeblood of Silicon Valley, allowing startups to access global talent at competitive rates.

The Policy Shift

The Trump administration, returning to power or influencing policy through legislative leverage, introduced a punitive fee structure. A new mandate required a $100,000 payment for every new employer-filed H-1B visa application submitted after September 21, 2025. Furthermore, strict "national interest" exemptions were narrowed, and denial rates for new petitions ticked upward to 2.8%.

The Impact on Startups

This policy created an immediate bifurcation in the market. Large incumbents like Microsoft, Google, and Nvidia could absorb a $100,000 fee as a rounding error in their hiring budget. Nvidia CEO Jensen Huang and OpenAI’s Sam Altman spoke "optimistically" about the move, perhaps recognizing that it acted as a moat, preventing smaller competitors from accessing the same talent pool.

For startups, however, this was an extinction-level event.

• The Cost of Talent: A seed-stage startup hiring five engineers suddenly faced a half-million-dollar up-front sunk cost. This destroyed the "runway" calculations of thousands of companies.

• The "Brain Drain": Startups that could not pay the fee were forced to rescind offers or rely on remote work, which introduced its own complexities regarding IP and data sovereignty.

• The "Gold Card" Bifurcation: While a "Gold Card" pathway was introduced for individuals gifting $1 million to the Department of Commerce, this was irrelevant to the rank-and-file engineers startups needed.

Startup Mortality Correlation:

Our analysis of the 60 Startup Failure List indicates that companies heavily reliant on distributed, international teams without a localized US presence were disproportionately represented in the failures. Unbabel (AI translation) and DefineCrowd (Data Labeling) faced "scaling pains" and "financial issues" that correlate with the increased friction of managing a global, visa-dependent workforce.

2.2 The "October 2025" Recession and Government Shutdown

The macroeconomic backdrop of late 2025 was defined by extreme volatility. While the broader S&P 500 remained relatively buoyant due to the heavy weighting of mega-cap tech, the "real economy" and the startup ecosystem entered a sharp contraction.

The Cleveland Fed's Warning

The Cleveland Federal Reserve's recession model flashed a warning signal, predicting a 24% probability that the US economy was already in a recession by October 2025. This quantitative signal was matched by qualitative distress in the B2B sector.

The Shutdown Paralysis

Compounding the economic slowdown was a prolonged US government shutdown in October 2025. This was not merely a political theater; it was a liquidity crisis for the "Deep Tech" sector.

• Frozen Capital: Thousands of startups in defense tech, climate tech, and biotech rely on government grants (SBIR/STTR) and contracts for non-dilutive funding. When the government shut down, the invoices stopped being paid.

• The "Valley of Death": Startups like LumenOrbit (Space Data)  found themselves in the "Valley of Death"; the gap between research funding and commercial revenue. The shutdown extended this valley beyond their cash reserves.

• Regulatory Stasis: The shutdown also paralyzed the Department of Labor, halting the certification of Labor Condition Applications (LCAs) required for H-1B filings, creating a double-bind with the new visa fees.

The Egregore's Purge:

The shutdown acted as a stress test. Companies with diversified revenue streams survived. Companies wholly dependent on government timelines, like Aura Labs (execution failure), were purged.

2.3 The Energy Crisis: The Hidden Tax on AI

The third executioner was the cost of energy. The explosion of Generative AI created an insatiable demand for electricity. By 2025, data center energy usage had surged to the point where "power is not a joke anymore".

The Cost of Compute

Startups building AI models faced a "Compute Trap."

1. Rising Demand: To stay competitive, they needed to train larger models or run more expensive inference.

2. Rising Cost: The cost of electricity; and thus the cost of cloud compute; rose significantly.

3. Broken Unit Economics: Startups like Stability AI (struggling) and Graphcore (bankrupt) could not pass these costs on to consumers who were accustomed to subsidized or free AI tools.

The "energy overhead" became a primary driver of broken unit economics, a leading cause of failure cited for Nook, Prophesee, and Northvolt.

Section 3: Case Studies in Failure

The theoretical framework and macroeconomic context provide the "why." The following case studies provide the "how." These narratives detail the specific trajectories of prominent failures, illustrating the relentless efficiency of the Market Egregore.

Case Study 1: Northvolt: The Green Giant that Starved

Sector: Battery Technology / Green Energy

Raised: $14 Billion+

Outcome: Chapter 11 (US) & Bankruptcy (Sweden), March 2025

Primary Cause: Geopolitical & Execution Misalignment

The Narrative:

Northvolt was the poster child of European industrial ambition. Founded by ex-Tesla executives, it promised to build the "world's greenest battery" and secure Europe's independence from Asian supply chains. It raised over $14 billion, including a massive $5 billion green loan in 2024. It had contracts with BMW, Volkswagen, and Volvo.

The Failure:

The collapse of Northvolt was a failure of scaling in a hostile geopolitical environment.

1. The Production Hell: The company's flagship gigafactory in Skellefteå, Sweden, was a disaster of execution. It targeted 16 GWh of production but managed only 1 GWh by the time the crisis hit. The "internal challenges" in ramping up highly complex manufacturing were underestimated.

2. The BMW Cancellation: In a crushing blow, BMW cancelled a €2 billion order in mid-2024 due to delivery delays. This was the first domino.

3. The Geopolitical Squeeze: Northvolt's business model assumed a certain price point for energy and raw materials. However, 2024-2025 saw "rising capital costs" and "geopolitical instability". The cost of raw materials (lithium, cobalt) remained volatile, while Chinese competitors like CATL continued to lower prices, aided by state subsidies and lower energy costs.

4. The Capital Trap: Building a gigafactory requires massive CapEx. When interest rates rose, the cost of servicing Northvolt’s debt exploded. The "Capital Strike" meant that when they needed a lifeline, the investors (Goldman Sachs, VW) balked, writing down their stakes instead of throwing good money after bad.

The Insight:

Northvolt failed because it tried to fight the Market Egregore's demand for efficiency with a political narrative of sovereignty. The market wanted cheap batteries; Northvolt offered expensive, "ethical" ones that it couldn't even deliver.

Case Study 2: Graphcore: The Hardware Betrayal

Sector: AI Hardware / Semiconductors

Raised: $700 Million+

Outcome: Acquired by SoftBank for ~$600M (estimated) in July 2024, effectively a wipeout for early common stock.

Primary Cause: Ecosystem Misalignment

The Narrative:

Graphcore was the UK's champion in the semiconductor wars. Valued at $2.8 billion in 2020, it built the Intelligence Processing Unit (IPU), a chip theoretically superior to Nvidia’s GPU for certain AI tasks.

The Failure:

Graphcore’s collapse illustrates the power of "Ecosystem Moats."

1. The CUDA Wall: Nvidia’s dominance is not just hardware; it is the CUDA software platform. Millions of developers code in CUDA. Graphcore required them to learn a new language (Poplar). The friction was too high.

2. The Revenue Collapse: In 2023, Graphcore posted zero revenue growth and pre-tax losses that widened. By the end of 2023, it had just $4 million in revenue against hundreds of millions in burn.

3. The Microsoft Loss: A landmark deal to supply processors to Microsoft Azure was scrapped, leaving Graphcore without a hyperscaler anchor client.

4. The Talent Exodus: In August 2025, co-founder Simon Knowles exited the company entirely, a year after the SoftBank acquisition, signaling the final capitulation of the original vision.

The Insight:

Graphcore bet on technological superiority (Level 1) but failed at Ecosystem Alignment (Level 3). The Market Egregore is path-dependent; once it selects a standard (Nvidia), it punishes deviation unless the improvement is 10x. Graphcore was perhaps 2x better, but 10x harder to use.

Case Study 3: Cerebras Systems: The IPO That Wasn't

Sector: AI Hardware / Wafer-Scale Computing

Raised: $1.1 Billion (Series G)

Outcome: IPO Withdrawal, October 2025

Primary Cause: Investor Relations & Client Concentration Risk

The Narrative:

Cerebras Systems, known for its dinner-plate-sized "Wafer Scale Engine," seemed poised to succeed where Graphcore failed. It raised a massive $1.1 billion Series G in late 2025, valuing the company at $8.1 billion. It filed for an IPO, expecting to be the blockbuster listing of the year.

The Failure:

On October 2025, Cerebras withdrew its S-1 filing.

1. Client Concentration: The S-1 revealed a terrifying statistic: 87% of its revenue came from a single client, G42 in the UAE.

2. The CFIUS Shadow: The Committee on Foreign Investment in the United States (CFIUS) had launched a probe into the G42 deal due to national security concerns regarding AI transfer to the Middle East (and potentially China).

3. The "Stale Data" Excuse: CEO Andrew Feldman claimed the withdrawal was because the S-1 was "stale" and didn't reflect the new funding. However, the timing; coinciding with the October 2025 mini-recession and the government shutdown; suggests the public markets refused to swallow the risk.

4. The Egregore's Rejection: The public market (the ultimate Egregore) rejected a company with such extreme dependency on a geopolitically risky client. Cerebras was forced to retreat to private markets (Series G) where terms could be dictated by insiders rather than the broader public.

Case Study 4: Nook: The "AI Wrapper" Cull

Sector: AI Sales Tech / Marketplace

Raised: $20 Million (Backed by Y Combinator)

Outcome: Shutdown

Primary Cause: Broken Unit Economics & Lack of Defensibility

The Narrative:

Nook promised to revolutionize sales with "AI Agents" that could automate prospect engagement. It was a classic YC darling, raising quick capital on the generative AI hype cycle.

The Failure:

Nook's failure was a classic case of the "Wrapper Problem."

1. Indistinguishable Product: Nook’s core value prop, AI dialing, and notes was quickly absorbed by the foundation models themselves (OpenAI, Salesforce).

2. The "Agent" Farce: Reports surfaced that the AI agents "excelled at having meetings... and had a lot of internal conflicts, and ended up with no product". It became a parody of automation; complexity without productivity.

3. Broken Economics: Nook suffered from "broken unit economics" and "valuation mismatch". The cost to acquire a customer (CAC) in a saturated market was astronomical, while retention (LTV) was low because the product had no moat.

4. Cultural Issues: The startup faced "cultural issues" and "execution failure," likely stemming from the pressure to scale a product that wasn't ready for prime time.

The Insight:

Nook was "Food for the Egregore." It served its purpose by testing a hypothesis (AI Sales Agents). The hypothesis failed (or was premature), and the market purged the resource to try again elsewhere.

Section 4: The "Capital Strike" and the Shift to Absolute ROI

The failures of 2025 cannot be understood without analyzing the behavior of the capital allocators. In 2023, VCs were playing a game of "Relative ROI"; trying to pick the best AI startup relative to others. By 2025, they shifted to "Absolute ROI" comparing startup returns against risk-free Treasuries and safe mega-caps.

4.1 The Series A Crunch

While Seed funding remained available (the "lottery ticket" phase), Series A and B funding dried up for companies that could not prove profitability.

• The Statistic: The conversion rate from Seed to Series A dropped significantly. VCs were no longer underwriting "growth"; they were underwriting "survival."

• The "Zombie" Phenomenon: This created a class of "Zombie Unicorns" companies like Stenn and EasyMile that had high valuations from 2021-2023 but no way to raise their next round. They didn't fail overnight; they starved slowly as the "Capital Strike" took hold.

4.2 The Rotation to Infrastructure

The Egregore directed capital away from applications (SaaS) and toward infrastructure (Energy, Data Centers, Chips).

• The Evidence: While Nook and Atrium died, Microsoft and BlackRock launched a $100 billion AI infrastructure fund. The market signaled that the value capture had moved to the physical layer.

• Tokenization: Simultaneously, capital began flowing into Real World Asset (RWA) Tokenization. Initiatives in Abu Dhabi (ADGM) and by major banks signaled a desire to make illiquid assets liquid. Startups that facilitated this (Fintech infrastructure) survived; startups that ignored it (traditional Fintech) struggled.

Section 5: The Hierarchy of Survival: A New Framework

Based on the forensic analysis of the 2023-2025 failures, we propose a new "Hierarchy of Survival" for the modern startup.

Table 1: The Hierarchy of Market Alignment (2025 Edition)

Key Finding:

Most startups focus on Levels 1 and 2. The failures of 2025 were predominantly at Levels 3, 4, and 5.

Conclusion: The Age of Awareness

The "90-99% failure rate" is not a constant of nature; it is a variable dependent on Awareness. The cohort of 2023-2025 failed because they were built for a world that no longer existed; a world of zero interest rates, open borders, and cheap energy.

They were executed by the Market Egregore because they violated the new Order.

• They ignored Policy: And the H-1B fees crushed them.

• They ignored Energy: And the compute costs starved them.

• They ignored Geopolitics: And the trade wars stranded them.

The Recommendation:

Future ventures must replace the "Move Fast and Break Things" ethos with a "Move Strategically and Align with Reality" doctrine. The role of the CEO is no longer just Chief Executive Officer; it is Chief Geopolitical Officer. They must track visa fees as closely as server costs. They must model energy shocks as closely as conversion rates.

The Market always wins. The only way to survive is to be on the side of the victory.

Disclaimer:

If you are using this publication for research purposes, or articles or related uses, ensure to accredit this publication and acknowledge Beunec as the conceptual origin. Readers must conduct their own research investigation before practical application; Beunec and the author disclaim responsibility for any errors or outcomes. By reading, you agree to these terms.

Works cited

1. The Order of Market Egregore by Olu Akinnawo {Internal Research Publication}.

2. Beunec Consulting LLC, & Peltier, I. (2025). Be five moves ahead: Maximize your decision-making strategies with untapped information. Independently published.

3. The Importance of ROI in Investment Appraisal - FasterCapital, accessed November 21, 2025, https://fastercapital.com/content/The-Importance-of-ROI-in-Investment-Appraisal.html

4. Northvolt files for bankruptcy in Sweden, accessed November 21, 2025, https://northvolt.com/articles/northvolt-files-for-bankruptcy-in-sweden/

5. Graphcore 2025 Company Profile: Valuation, Investors, Acquisition | PitchBook, accessed November 21, 2025, https://pitchbook.com/profiles/company/166649-68

6. Graphcore cofounder exits company one year on from SoftBank acquisition | Sifted, accessed November 21, 2025, https://sifted.eu/articles/graphcore-cofounder-exits-company-one-year-on-from-softbank-acquisition

7. How Trump's H-1B Reform Could Harm American Tech Innovation - Time Magazine, accessed November 21, 2025, https://time.com/7319507/h1b-trump-tech-innovation-ai/

8. US tech giants dominate H-1B visa approvals in 2025 as Indian firms scale back, accessed November 21, 2025, https://m.economictimes.com/nri/work/us-tech-giants-dominate-h-1b-visa-approvals-in-2025-as-indian-firms-lose-ground/articleshow/125428036.cms

9. Exclusive: Lawmakers push back against Trump's H-1B visa fee, accessed November 21, 2025, https://liccardo.house.gov/media/in-the-news/exclusive-lawmakers-push-back-against-trumps-h-1b-visa-fee

10. Are H-1B Changes 'A Strategic Opening' For Startup Creation Or An 'Insurmountable Founder Tax'? - Crunchbase News, accessed November 21, 2025, https://news.crunchbase.com/business/h-1b-changes-startup-founder-impact-cost-o1-visa-entrepreneur-alternatives/

11. Forecasting US Recessions in Real-Time Using Regional Economic Sentiment, accessed November 21, 2025, https://www.clevelandfed.org/publications/economic-commentary/2025/ec-202513-forecasting-us-recessions-in-real-time-using-regional-economic-sentiment

12. How the U.S. Government Shutdown Impacts Startups - Tomorrow Access, accessed November 21, 2025, https://tomorrowaccess.com/en/news/17968

13. Crisis Ahead: Power Consumption In AI Data Centers - Semiconductor Engineering, accessed November 21, 2025, https://semiengineering.com/crisis-ahead-power-consumption-in-ai-data-centers/

14. The Impending Wave: Artificial Intelligence Startups Facing Bankruptcy Risks - Mejeticks, accessed November 21, 2025, https://www.mejeticks.com/post/ai-startup-bankruptcies-gpu-costs-monetization

15. Northvolt Files for Bankruptcy - ESG Today, accessed November 21, 2025, https://www.esgtoday.com/northvolt-files-for-bankruptcy/

16. Northvolt's Bankruptcy: How Does It Impact Europe's Battery Industry? - CarbonCredits.com, accessed November 21, 2025, https://carboncredits.com/northvolts-bankruptcy-how-does-it-impact-europes-battery-industry/

17. Cerebras Systems Pulls Plug On Its IPO Days After Big Fundraise - Crunchbase News, accessed November 21, 2025, https://news.crunchbase.com/public/ipo/cerebras-systems-halts-ipo-ai-semiconductors/

18. Cerebras CEO explains why IPO was pulled - The Register, accessed November 21, 2025, https://www.theregister.com/2025/10/06/cerebras_ipo_rest/

19. Cerebras Pulls Its IPO Days After $1.1 Billion Round - AIM Media House, accessed November 21, 2025, https://aimmediahouse.com/market-industry/cerebras-pulls-its-ipo-days-after-1-1-billion-round

20. The all-AI startup that failed: A cautionary case study - The American Genius, accessed November 21, 2025, https://theamericangenius.com/the-all-ai-startup-that-failed-a-cautionary-case-study/

21. Billions flow into AI in 2025: how Big Tech is rewriting the rules of venture capital | Vestbee, accessed November 21, 2025, https://www.vestbee.com/insights/articles/state-of-ai-in-2025-how-big-tech-is-rewriting-the-rules-of-venture-capital

22. Tokenization of Financial Assets - IOSCO, accessed November 21, 2025, https://www.iosco.org/library/pubdocs/pdf/IOSCOPD809.pdf

23. Asset tokenization now and in the future - State Street, accessed November 21, 2025, https://www.statestreet.com/tw/en/insights/digital-digest-december-2024-tokenization-and-future