Is Brazil Ready for Digital Independence? Inside the New Sovereign Cloud Plan

Brazil’s so called “Brazilian Cloud” is planned to run on infrastructure owned by the public sector within Brazil. It will be run by a state company. This was described by people who took part in a public consultation led by government technology offices, and It is something that worries many Brazilians.

The goal is to see how digital control works in real life. Organizers say they want more than just promises from vendors. One idea for a check is called the Disconnection Test. Under that test, links to the outside internet would be cut. Then evaluators would check whether key services still work. This includes login and trust systems, APIs, virtual machines, Kubernetes, storage, databases, key handling, monitoring, backup work, and admin tasks.

Serpro, a federal data processing firm tied to the effort, said its testing work has grown a lot. It started at about 70 checks and now is close to 500. In some cases, paper reviews implied that rules were met. But hands on trials later showed gaps.

The government says it will look into how much it relies on AI tools, both hardware and software. That includes NVIDIA’s CUDA setup. The team plans to review the whole stack and check if parts can be swapped out while the platform keeps working. For now, building chips is not the main goal. This early phase is about the software side.

Officials expect the next cloud to offer AI and machine learning services. This would cover generative AI and large language models. It would also include options for tuning models, putting them into use, and scaling them up. The plan also calls for data traceability and model checks for use in the public sector.

They propose a “Sovereignty Gradient.” In that approach, 30% of the score would go to data sovereignty. 40% would go to operational sovereignty. The last 30% would go to technological sovereignty. Vendors would have to share details on their tech and control layers. They would also need to name who runs what, how updates are handled, and what licensing terms apply. The disclosure would include how outside users can access systems, what telemetry is sent, and whether parts can be replaced.

The work is coordinated by the Ministry of Management and Innovation, Serpro, the Brazilian Development Bank (BNDES), and the Brazilian Agency for Industrial Development (ABDI). Officials noted that the current consultation is early. It is not meant to pick a supplier. Instead, it aims to inform later technical, legal, and business options.

This effort also seeks to use public buying power to boost local tech know-how. If intellectual property is created in a later partnership, it must be shared. In addition, knowledge transfer to teams at public companies is expected, along with research and development carried out in Brazil.

Officials said the contract is likely to have a reversibility clause. If a private partner pulls out, stops work, or ends the deal, the state would have to run the platform on its own. That would include getting permanent access to the source code.

Officials also said the effort is not meant to wipe out foreign tools or overseas vendors. The plan is to spot where the project depends on outside tech. Then the goal is to make sure those dependencies do not block the government from using the system. If needed, they would switch to other options or swap out suppliers.

The key issue is simple. Can Brazil keep its digital infrastructure running if a supplier, a technology provider, or a foreign country stops support?

Is an AI Bubble Next? Comparing Today's Tech Boom to the 2008 Financial Crisis

Recent analyses suggest a potential economic downturn, possibly more severe than the 2008 subprime mortgage crisis, driven by the overvaluation of leading technology companies, often dubbed the "Magnificent Seven," and speculative investments in artificial intelligence (AI). 

Since the launch of ChatGPT in late 2022, the S&P 500 has surged about 90%, with most gains driven by a small group of AI-linked technology giants, including Microsoft, Apple, Alphabet, Nvidia, and major data-center operators.

Nvidia has emerged as the sector’s standout, evolving from a gaming chipmaker into a central supplier of AI infrastructure and approaching record-breaking market valuations. However, critics warn that much of the investment flowing into AI companies is being recycled within the sector itself, creating a tightly interconnected financial system that could amplify risks if sentiment shifts.

This perspective is underscored by concerns from industry leaders, including CEOs of major tech firms, who hint at the fragility of current market valuations and the potential for widespread economic fallout.

High-profile investors have also entered the debate. Michael Burry, known for predicting the 2008 financial crisis, has publicly bet against the AI boom, arguing that extreme capital concentration often precedes major downturns. His warnings have prompted some fund managers to reduce exposure to technology stocks. Critics, however, note that several of Burry’s post-2008 predictions did not come true.

Regulators and analysts also have raised red flags. The Bank of England has cautioned that AI-related stocks may be overvalued, while media reports highlight soaring executive and researcher compensation as a sign of overheating. Despite massive funding, key players such as OpenAI are not yet profitable.

 

Echoes of the Subprime Meltdown 

The 2008 subprime crisis serves as a critical precedent for understanding the current anxieties. At its core, the subprime crisis was fueled by an immense creation of fictitious capital within the U.S. real estate market. This involved inflated property values, often detached from their intrinsic worth, and a pervasive system of securitization. 

The Subprime Mechanism: 

• Fictitious Capital: The housing bubble led to assets being priced unrealistically. 
• Securitization (CDOs): Mortgage-backed securities, known as Collateralized Debt Obligations (CDOs), were widely distributed globally. These instruments, similar to Brazil's Real Estate Receivable Certificates (CRIs), allowed banks to offload risk by selling debt to investment funds worldwide. 
• Excess Liquidity and Risky Lending: An abundance of capital in the financial system led banks to extend credit to increasingly unqualified borrowers, including those with no ability to repay, in pursuit of higher returns. This was rationalized by a booming market where property values and rents were consistently rising, seemingly ensuring repayment. 
• Bubble Burst: The unsustainable rise in property prices eventually led to a saturation point, with properties becoming vacant and rents failing to cover mortgage payments. Defaults surged, leading to foreclosures and a rapid decline in property values as seized assets flooded the market. 
• Global Contagion: The failure of CDOs caused investment funds holding these securities to lose massive value, triggering a liquidity crisis. Investors rushed to redeem funds, forcing asset liquidations across various markets (stocks, bonds), creating a domino effect that crippled the global financial system. 

The Current AI and Tech Bubble Today, concerns center on the Magnificent Seven (The largest tech companies in the S&P 500) which disproportionately drive market growth. The AI sector, in particular, is seen as a new locus of fictitious capital formation. Despite massive investments, AI technologies are not yet generating sufficient revenue to justify their soaring valuations, drawing parallels to the dot-com bubble of the late 1990s. 

Industry figures, such as Microsoft's CEO and Sam Altman (OpenAI), have openly acknowledged the existence of this bubble, even suggesting that government intervention might be necessary should it burst. This indicates an awareness within the industry that current valuations are unreal and predicated on future cash flows that may not materialize for many companies. 

Factors Contributing to the Current Bubble: 

• Unjustified Valuations: Companies like Palantir, trading at 116 times revenue, exemplify valuations detached from fundamental asset value, which theoretically should be based on the ability to generate future cash flow. 
• Passive ETF Management: Over half of the capital entering the U.S. stock market is managed passively through algorithms that automatically buy index proportions. This mechanism artificially inflates the prices of larger companies, with studies suggesting that Apple's price, for instance, was inflated by 23% due to these passive flows alone. 
• Baby Boomer Effect: The impending retirement of the baby boomer generation is expected to lead to significant withdrawals from investment funds, potentially reversing the positive inflow trends and exacerbating market instability. 

Potential Impact of a Bursting Bubble 

Should this bubble burst, the consequences are projected to be severe and systemic. The disappearance of wealth would lead to a sharp reduction in consumer spending and overall economic activity. Assets, widely used as collateral throughout the financial system, would trigger cascading losses, leading to a profound liquidity crisis and a halt in money circulation. 

Historically, such crises result in the failure of smaller businesses, the contraction of medium-sized enterprises, and the opportunistic acquisition of undervalued assets by larger entities. This process invariably leads to a further concentration of wealth among the already rich and an expansion of poverty, illustrating a cyclical aspect of capitalism where crises of overproduction of fictitious capital are periodically resolved through its destruction.

Analysts also point to broader consequences, including environmental concerns tied to the rapid expansion of energy-intensive data centers. 

For now, experts agree on one point: bubbles are only confirmed after they burst, but the warning signs are becoming harder to ignore.