Hugging Face is often described as the “GitHub of AI.” Its platform allows developers and companies to share, download and build on AI models and datasets. The startup was valued at $4.5 billion in 2023, meaning Nvidia is paying nearly three times that valuation.

Why Nvidia Wants Hugging Face

The deal goes beyond simply buying another AI company. Nvidia wants open-source AI to remain a strong alternative to closed models from OpenAI and Anthropic.

That matters because major closed-model developers are increasingly working on their own AI chips to reduce dependence on Nvidia. Supporting a large ecosystem of open models, many of which run on Nvidia hardware, could help protect demand for the company’s GPUs. Nvidia has also been investing heavily in its own Nemotron open models.

The price is striking given Hugging Face’s relatively small business. The company generates only about $150 million in annual revenue, according to Reuters, putting the $12.9 billion purchase price at roughly 86 times revenue.

There is also a potential challenge. Hugging Face has built its reputation as a relatively neutral platform supporting models and hardware from across the AI industry, including Nvidia competitors such as AMD and Intel. Nvidia ownership could raise questions about whether that neutrality can continue.

The post Nvidia Agrees to Buy Hugging Face for $12.9 Billion in Major AI Deal first appeared on Finblog.

Unlike the US, where much of the AI race has focused on building the most powerful models, China has aggressively pushed companies to actually use the technology. The share of Chinese industrial companies using AI models and agents jumped from 9.6% in 2024 to 47.5% last year, according to IDC.

The impact is already visible. AI is being used across programming, translation, media production and logistics, while robots are increasingly appearing in parcel delivery, food delivery and other physical jobs. China’s live-action short-video production reportedly fell about 75% year over year in the first quarter, as generative AI became more widely used in content creation.

For workers, that efficiency comes with a cost. One Beijing programmer interviewed by AP was laid off alongside around 160 colleagues, while a translator said industry pay has fallen by more than half compared with previous years. Chinese technology companies have also cut or restructured tens of thousands of positions, partly as AI changes how work is done.

China Has a Smaller Safety Net

The risk is particularly important because China’s public social spending is relatively low compared with major developed economies. China spends around 10% of GDP on public social programs, compared with 23% in the US, 27% in Germany and 32% in France, according to OECD data highlighted by Semafor.

China is already struggling with youth employment. Overall urban unemployment is around 5%, but unemployment among 16-to-24-year-olds who are not students is roughly three times higher. Economists warn that rapid AI adoption could worsen those pressures and potentially weigh on consumer spending and social stability.

There is a longer-term upside. China’s population is aging and shrinking, and by 2050 there are projected to be fewer than two working-age adults for every retiree. Automation could eventually help the country compensate for that shrinking workforce.

The post Workers in China worry over being replaced as they adapt to the growing impact of AI on jobs first appeared on Finblog.

According to Axios, communities are increasingly pushing back against new data center projects over concerns about electricity costs, water consumption, noise and pressure on local infrastructure. That frustration is now moving into election campaigns ahead of the US midterms.

Public attitudes toward AI itself are also becoming more cautious. A Pew Research Center survey cited by Axios found that 52% of Americans are more concerned than excited about AI, up significantly from 2021.

The political shift is already visible:

• In Wisconsin, Republican gubernatorial candidate Tom Tiffany is attacking his Democratic opponent over plans to turn the state into a major AI and data center hub.
• Pennsylvania Governor Josh Shapiro has introduced tougher rules for data centers, including requirements that developers cover their own energy costs and source at least 10% of their electricity from clean energy.
• In Colorado, some communities are introducing restrictions or temporary pauses on new projects as officials consider their impact on electricity and water supplies.

The stakes are high for companies such as OpenAI, Microsoft, Meta, Amazon and Google, which need enormous amounts of computing capacity to expand their AI businesses. More than 4,000 data centers are already operating across the US, with over 3,000 additional facilities planned, according to a separate Axios report.

The AI industry is now trying to change the conversation by emphasizing the jobs, investment and economic growth these projects can bring to local communities. But as data centers become more visible in everyday life, the industry’s infrastructure expansion is becoming harder to separate from politics.

Investor takeaway: AI demand remains strong, but building the physical infrastructure behind it is becoming more complicated. If local opposition leads to slower approvals, tougher regulations or higher energy costs, data center development could become a new bottleneck for the AI boom.

Source: Axios

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

The post Behind the Curtain: The new existential threat to AI first appeared on Finblog.

The concern is simple: if future AI electricity demand falls short of today’s ambitious forecasts, investors could be left with expensive power plants that are no longer needed. Semafor points to the early 2000s, when investors lost heavily after building power capacity for demand that ultimately failed to materialize.

Instead, some private equity firms are turning toward regulated utilities, where returns are usually lower but much more predictable. In states such as Louisiana and Florida, utilities operate as regulated monopolies and electricity prices are determined through a formal regulatory process.

The AI boom is creating an unusual opportunity to buy these assets. Utilities including Duke Energy and AEP need billions of dollars to fund new infrastructure, pushing some companies to sell non-core regulated businesses to raise cash. Bernhard Capital Partners says it has already completed six regulated gas and power utility acquisitions in two years.

Money is flowing quickly into the sector. Global private equity investment in utilities exceeded $69 billion in 2025, up 50% from the previous year. Much of the new AI power infrastructure is expected to rely on natural gas, bringing additional regulatory and environmental scrutiny.

Investor takeaway: Instead of betting directly on which AI company or data center wins, investors are increasingly looking at the utilities supplying the electricity. The returns may be less spectacular, but regulated power businesses could offer a more predictable way to benefit from AI’s growing energy demand.

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

The post AI Investors Are Finding a Safer Way to Bet on the Data Center Boom first appeared on Finblog.

CoreWeave remains one of the clearest examples. Q2 revenue jumped 112% to $2.6 billion, with committed contracts accounting for 98% of revenue. Its backlog reached $104 billion, up 246%, even before another $25 billion in customer commitments signed early in Q3.

But that growth is expensive. CoreWeave spent $9.4 billion on CapEx in Q2 and recorded a $626 million net loss, including $640 million in interest expense. It has now raised its 2026 CapEx forecast to $35 billion to $39 billion as it works toward more than 1.85 GW of active power by year-end.

Nebius is growing even faster. Revenue surged 454% to $582 million, with its AI Cloud generating $575 million, or 98% of total revenue. Gross margin reached 77%, while adjusted EBITDA came in at $236 million, a 41% margin.

Its Q2 CapEx, however, reached $5.7 billion, almost ten times quarterly revenue. Nebius expects to spend $20 billion to $25 billion this year, although customers are helping fund the expansion. More than $9 billion in customer prepayments are expected in 2026, covering roughly 50% to 60% of related CapEx. New contracts are also becoming more profitable, with estimated payback periods falling to around 22 months.

Cerebras is taking a different route by using its own AI processors rather than relying on Nvidia GPUs. Q2 revenue rose 74% to $180 million, while Cloud & Other Services jumped 281% to $126 million. Its remaining performance obligations reached $25.4 billion, showing that demand is running well ahead of current revenue.

The common challenge is clear: AI demand is not the problem. Funding the infrastructure needed to meet that demand is.

Investor takeaway: Neoclouds are benefiting from a genuine shortage of AI computing capacity, but revenue growth alone will not determine the winners. The companies that can build capacity quickly while controlling debt, improving margins and generating better returns on each new data center are likely to have the strongest long-term position.

Source: App Economy Insights

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

The post AI Compute Demand Is Booming, but Neocloud Growth Comes at a Huge Cost first appeared on Finblog.

Modern AI agents are very different from traditional automated bots. Instead of following a fixed script, they can understand a webpage, decide what to click, react to errors and change their strategy until they complete a task. That makes simply labeling a visitor as either “human” or “bot” increasingly difficult.

CAPTCHAs are also losing effectiveness. A 2026 study cited by Tech Scoop found that some commercial CAPTCHA-solving services achieved near-perfect bypass rates, with costs as low as roughly $0.10 per 1,000 challenges.

The problem is becoming more complicated because not all AI traffic is unwanted. Companies may actually want AI shopping assistants, travel agents and other automated services to access their websites and perform tasks for users.

As a result, cybersecurity could shift from asking “Is this a bot?” to asking whether a particular action can be trusted. Websites could increasingly evaluate a combination of identity, authorization, browser history and behavior throughout a session instead of relying on a single CAPTCHA.

The transition has already reached standards organizations. NIST launched an AI Agent Standards Initiative in February 2026, while other groups are exploring ways for websites to identify AI agents and determine exactly what they are authorized to do.

Bottom line: CAPTCHAs probably will not disappear immediately, but AI is making them much weaker as a standalone defense. The next generation of web security may focus less on proving someone is human and more on proving that an AI agent is identifiable, authorized and behaving as expected.

Source: Tech Scoop

The post CAPTCHAs Are Failing as AI Agents Become Harder to Detect first appeared on Finblog.

According to Bloomberg, more investors are using AI to write code, test trading strategies and automate buying and selling decisions. AI assistants have dramatically lowered the technical barrier, allowing people with little programming experience to experiment with algorithmic trading.

But building a profitable strategy remains much harder than building the software itself.

Bloomberg highlights the story of Joel Rieger, a software sales executive who spent more than a year developing an automated options trading system. He wrote Python code, tested hundreds of strategies and analysed hundreds of stocks. In the end, he found that his results were no better than simply investing in an S&P 500 index fund, leading him to abandon the project.

His experience reflects a broader reality. While AI can help traders create sophisticated tools in a fraction of the time, it cannot guarantee better investment returns. Professional hedge funds still have major advantages, including access to proprietary data, faster infrastructure and teams of quantitative researchers.

Even so, interest continues to grow as AI makes algorithmic trading more accessible and affordable for retail investors.

Investor takeaway: AI is changing how retail investors approach the market, making it easier to build trading systems. But successful investing still depends on having a profitable strategy, not just better technology.

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

The post AI Is Turning Retail Traders Into DIY Hedge Funds first appeared on Finblog.

The report argues that improvements in AI efficiency are encouraging companies to build and deploy more AI applications rather than spend less on computing. As a result, demand for GPUs, memory chips, networking equipment and data centre capacity continues to grow faster than efficiency gains can offset.

This reflects the Jevons Paradox, an economic principle suggesting that when a technology becomes more efficient, overall usage often increases instead of declines.

Several factors are expected to keep AI infrastructure expensive:

• Rising prices for high-bandwidth memory (HBM), a key component in AI chips.
• Limited availability of power and data centre capacity.
• Rapid growth in AI inference, as businesses move from training models to deploying AI products at scale.
• Continued heavy investment from hyperscalers such as Microsoft, Amazon, Google, and Meta, which are expanding AI infrastructure to meet growing demand.

The report notes that while newer AI models require fewer resources to train, they are also making AI more accessible. That is encouraging businesses to launch more AI-powered services, increasing overall computing demand instead of reducing it.

This trend is also changing where spending is directed. While training the largest frontier models remains expensive, inference, the process of running AI models for millions of users, is becoming an even bigger driver of infrastructure investment.

Investor takeaway: More efficient AI models are unlikely to reduce industry spending. Instead, lower costs per task could accelerate AI adoption, supporting continued demand for chips, memory, networking equipment and data centres. That means companies supplying AI infrastructure may continue to benefit, even as AI models themselves become cheaper to run.

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

The post AI compute could become more expensive even as GPUs improve first appeared on Finblog.

AI-related companies have become a particular target for short sellers, with investors increasingly questioning whether massive capital spending on AI infrastructure will translate into meaningful profits.

The latest earnings season is expected to test that thesis.

Alphabet, which reported quarterly results on Wednesday, came under pressure after investors focused on its rising AI spending and delays to its latest Gemini model. At the same time, rapid progress in China’s open-source AI models has raised concerns that US tech giants may be overinvesting in data centres and computing capacity.

Investors are also watching hyperscalers closely as AI infrastructure costs continue to climb, putting pressure on margins despite strong revenue growth.

According to The Atlantic, AI-related companies now carry a combined valuation of roughly $27 trillion, increasing pressure on the sector to prove that heavy investment will eventually generate sustainable profits. If those expectations are not met, analysts warn the market could face a significant correction.

Investor takeaway: AI remains one of the market’s biggest long-term growth themes, but investors are becoming less willing to reward spending alone. Upcoming earnings from major technology companies will be closely watched for evidence that AI investment is translating into stronger profits rather than simply higher capital expenditures.

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

The post Investors Increase Bearish Bets on AI Stocks as Spending Concerns Grow first appeared on Finblog.

Former New York City mayor Michael Bloomberg criticized proposals for the government to take ownership stakes in AI companies, an idea backed by Sen. Bernie Sanders and reportedly discussed by President Donald Trump.

Writing in an opinion piece, Bloomberg argued that government ownership would discourage innovation, quipping, “Somewhere, Karl Marx is smiling.”

Meanwhile, another debate has emerged over the rapid progress of Chinese AI models.

The discussion intensified after Beijing-based Moonshot AI released its new open-source model, Kimi K3, prompting questions about whether the US should place tighter restrictions on AI technology developed in China.

David Sacks, who previously served as the White House’s AI and crypto czar, said Washington should clearly communicate any genuine security risks associated with Chinese AI models and regulate them where necessary. However, he also cautioned against creating unnecessary fear to discourage their use.

His comments echoed concerns raised by OpenAI’s Dean Ball, who warned against what he described as a growing “FUD” (fear, uncertainty, and doubt) approach to AI policy.

At the same time, Bloomberg News reported that the Trump administration is considering establishing an independent AI watchdog within the Securities and Exchange Commission (SEC). The proposed body would focus on AI-related risks in financial markets and corporate disclosures.

For investors, the debate highlights that AI regulation is becoming just as important as AI innovation itself. Future rules on competition, national security, and corporate oversight could shape how AI companies operate, compete, and attract investment in the years ahead.

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

The post AI Regulation Debate Heats Up as Bloomberg, Sacks, and Trump Weigh In first appeared on Finblog.