From Wall Street to Silicon Valley, Anthropic Steals All the Spotlight from OpenAI

marsbitPublished on 2026-04-13Last updated on 2026-04-13

Abstract

From Wall Street to Silicon Valley, Anthropic is seizing the spotlight from OpenAI. In just one year, the power dynamics in the AI have shifted significantly. Anthropic is now challenging OpenAI across multiple fronts: market share, secondary market valuation, venture capital sentiment, and public perception. At the recent HumanX AI conference, the consensus was clear—Anthropic is the new darling of Silicon Valley. Its annualized recurring revenue (ARR) has reportedly reached $300 billion, surpassing OpenAI's $250 billion. In the secondary market, Anthropic's valuation has overtaken OpenAI's, with strong investor preference for its shares. Anthropic dominates the enterprise sector, holding 42-54% of the code generation market and 40% of the enterprise agent market, compared to OpenAI's 21% and 27%, respectively. It also leads in new enterprise adoption and cost efficiency. While OpenAI retains a strong consumer user base with ChatGPT, it faces challenges inization and high operational expenses. A leaked internal memo from OpenAI identified Anthropic as its biggest threat, emphasizing its compute infrastructure advantage, but the very need for such a memo highlights its defensive position. Despite OpenAI's strong backing from Amazon and NVIDIA, the market is now valuing efficiency, cost-effectiveness, and precise market fit—areas where Anthropic currently leads. However, experts caution that the AI race is far from over and the landscape remains highly fluid.

Written by: Dong Jing

In just one year, the power balance in the AI industry has quietly shifted. OpenAI, once a dominant force in the investment world, is now facing a comprehensive challenge from Anthropic—from enterprise market share to secondary market valuation, from venture capital reputation to social media buzz, Anthropic is eroding OpenAI's leading position in almost every dimension.

This shift in sentiment was amplified to the extreme at the HumanX AI Conference held in San Francisco last week. According to Business Insider, the attending venture capitalists and entrepreneurs almost reached a unified consensus: Anthropic is the new darling of Silicon Valley.

Roseanne Wincek of Renegade Partners stated bluntly, "Last year in Las Vegas, it felt like OpenAI was the clear winner, but now Anthropic seems to be several steps ahead." Meanwhile, Anthropic announced its annualized recurring revenue (ARR) has surpassed $30 billion, exceeding OpenAI's previously announced $25 billion, making it the world's highest-revenue AI unicorn.

The flow of market capital speaks volumes. According to Bloomberg, Anthropic's valuation on the secondary market has surpassed OpenAI's. Data circulating on social platforms shows that Anthropic's private market valuation is approximately $863.6 billion, while OpenAI's is about $846.1 billion. Furthermore, OpenAI's secondary shares are experiencing unprecedented cold reception, while buyers are queuing up to purchase Anthropic's shares.

Silicon Valley's Sudden Shift: Anthropic Becomes the Conference Protagonist

This year's HumanX conference was twice the size of last year's, with about 6,700 attendees paying over $4,000 per ticket. However, in stark contrast to last year's atmosphere in Las Vegas where everyone was betting on OpenAI, inside San Francisco's Moscone Center this year, Anthropic was the center of conversation.

Jared Quincy Davis, founder and CEO of AI cloud platform Mithril, said,

"They (Anthropic) have strong momentum. It's evident that their focus on the enterprise market, cutting-edge capabilities, and code generation, while deliberately avoiding some consumer scenarios, are all the right decisions."

Halfway through the conference, Anthropic released its latest model, Mythos, stating it was so powerful that it would not be released to the public due to cybersecurity risks, instead being made available only to select enterprises through a new initiative called "Project Glasswing." Tomasz Tunguz, founder and general partner at Theory Ventures, commented, "The Mythos model is significant; there is tremendous excitement in the market."

In contrast, it was almost impossible to find public supporters for OpenAI at the conference. Attendees' doubts about OpenAI centered on two points: the perplexing acquisition of the internet talk show TBPN, and the controversy surrounding CEO Altman's dealings with the Pentagon. Former Coatue and Kleiner Perkins partner Andy Chen stated, "A significant number of people have objections to Altman and his actions," and predicted a talent exodus from OpenAI.

Secondary Market: Unwanted Shares and Valuation Reversal

The signals from the capital market are more direct. As pointed out in a Wall Street Seen article, Bloomberg reported that in early April, six OpenAI institutional shareholders—including hedge funds and well-known VCs—attempted to sell approximately $600 million worth of OpenAI shares through the secondary market platform Next Round Capital. However, the platform contacted hundreds of institutional buyers, and not a single one was willing to take them.

Ken Smythe, founder of Next Round Capital, stated bluntly, "We simply couldn't find a single institutional investor willing to take these shares, and we have resources with hundreds of institutions." He also revealed that buyers told him they had $2 billion in cash, waiting solely to buy Anthropic stock.

This scenario is playing out on other trading platforms as well. On SPV trading platforms like Augment and Hiive, the trend of investors flocking to Anthropic and shunning OpenAI is already evident. Adam Augment, co-founder of Augment, stated, everyone believes Anthropic's valuation can catch up to OpenAI's, so they all want to buy in as soon as possible.

Valuation data confirms this judgment. Reports indicate that OpenAI's secondary market trading valuation is approximately $765 billion, a discount of about 10% from its last funding round valuation; whereas Anthropic's secondary market trading valuation has reached $600 billion, a premium of over 50% compared to its last funding round. The latest data circulating on social platforms shows that Anthropic's private market valuation has slightly surpassed OpenAI's.

The actions of Wall Street investment banks are equally telling. It is reported that several investment banks, including Morgan Stanley and Goldman Sachs, have begun pitching OpenAI shares to high-net-worth clients without charging performance fees; whereas Goldman Sachs still charges clients investing in Anthropic the customary 15% to 20% profit share.

B2B Market: Anthropic Has Formed an Overwhelming Advantage

Behind the revenue numbers, deeper structural changes are occurring in the enterprise market.

In the currently most important B2B sector for AI large models—code generation—Anthropic's Claude model already holds 42% to 54% of the global market share, while OpenAI has only 21%. In the enterprise agent market, Anthropic's share is 40%, OpenAI's is 27%.

Incremental data better illustrates the trend. Ramp data indicates that among enterprises newly procuring AI services in March 2026, a whopping 65% chose Anthropic, while only 32% chose OpenAI. By April 2026, Anthropic had over 1,000 enterprise customers with annual consumption exceeding $1 million, doubling in the past two months. API calls and enterprise customization services now account for over 80% of its total revenue.

The gap in cost efficiency is equally staggering. According to Wall Street Journal estimates, OpenAI's annual training costs will reach $125 billion by 2030, while Anthropic will need only about $30 billion—a difference of over four times. With its high revenue growth, Anthropic could achieve cash flow breakeven by 2027, while OpenAI's profitability timeline remains distant.

In contrast, OpenAI's only area of overwhelming dominance is the consumer side—ChatGPT currently has over 900 million weekly active users, but over 98% are free users, consuming vast amounts of computing power while generating almost no revenue. In February 2026, OpenAI attempted to introduce ads in ChatGPT, sparking widespread controversy.

On social platforms, user @deedy pointed out that OpenAI/ChatGPT has now begun placing bid ads on Claude-related keywords, remarking on how "the tables have turned." The post quickly gained significant attention.

OpenAI's Counterattack: Compute Advantage and Leaked Memo

Facing external doubts, OpenAI sent a confidential memo to shareholders this week, which was subsequently leaked. In the memo, OpenAI characterized Anthropic as its biggest competitive threat, while emphasizing its own leading advantage in compute infrastructure.

According to the memo's contents, OpenAI had 1.9 gigawatts (GW) of compute capacity in 2025, expects to increase this to the low double digits next year, and reach approximately 30 GW by 2030; in comparison, OpenAI estimates Anthropic had only 1.4 GW at the end of 2025 and expects to reach 7-8 GW next year. "Even by the highest estimates, our expansion pace is substantively leading and the gap continues to widen," the memo stated.

On Anthropic's side, it has agreements with Google and Broadcom to secure 5 GW of next-generation TPU compute power support starting in 2027.

However, the very leak of this memo has, to some extent, exposed OpenAI's defensive posture. The fact that a company once seen as the absolute industry leader now needs to specifically write to its shareholders explaining why it remains competitive is a signal telling enough in itself.

This Race is Far From Over

Despite Anthropic's strong momentum, several attendees cautioned against drawing conclusions too early. "These things change so fast," said Roseanne Wincek, "OpenAI could come back." Tomasz Tunguz also noted, "You wake up every day, and something has changed substantively."

OpenAI's fundraising capability remains strong. In its latest completed $122 billion financing round, Amazon contributed $50 billion and Nvidia $30 billion. Both are not purely financial investors but have locked in strategic returns through compute supply and cloud service contracts.

But the market's judgment has undergone a structural shift. The logic determining the outcome of the AI race is shifting from "who raises the most money, who has the grandest narrative" to "who can create value for users with the lowest cost, highest efficiency, and most precise market positioning." Under this new logic, Anthropic currently holds the upper hand.

Related Questions

QWhat are the key areas where Anthropic is challenging OpenAI's dominance according to the article?

AAnthropic is challenging OpenAI in multiple dimensions, including enterprise market share, secondary market valuation, venture capital sentiment, and social media buzz. It has surpassed OpenAI in annual recurring revenue (ARR), with $300 billion compared to OpenAI's $250 billion, and leads in code generation and enterprise agent market share.

QHow did the perception of Anthropic and OpenAI differ at the HumanX AI conference compared to the previous year?

AAt the previous year's conference in Las Vegas, OpenAI was seen as the clear winner. However, at this year's HumanX AI conference in San Francisco, Anthropic became the central topic of discussion, with attendees viewing it as the new Silicon Valley darling, while OpenAI faced criticism and a lack of public support.

QWhat does the secondary market activity reveal about investor preference between Anthropic and OpenAI?

ASecondary market activity shows a strong preference for Anthropic over OpenAI. OpenAI's shares faced unprecedented cold reception, with no institutional buyers for a $600 million offering, while investors were eager to buy Anthropic's shares, with some having $2 billion ready to invest. Anthropic's secondary market valuation also saw a premium of over 50%, while OpenAI's traded at a discount.

QIn which specific enterprise markets does Anthropic hold a significant advantage over OpenAI?

AAnthropic holds a significant advantage in the enterprise market, particularly in code generation, where it captures 42% to 54% of the global market share compared to OpenAI's 21%, and in the enterprise agent market, where it has a 40% share versus OpenAI's 27%. Additionally, 65% of new enterprise AI service purchases in March 2026 chose Anthropic.

QHow does OpenAI defend its competitive position against Anthropic in the leaked memo?

AIn the leaked memo, OpenAI highlights its compute infrastructure advantage, stating it had 1.9 GW of computing capacity in 2025, expected to grow to low double digits next year and reach about 30 GW by 2030. It estimates Anthropic had only 1.4 GW at the end of 2025 and projects it will reach 7-8 GW next year, emphasizing that OpenAI's expansion pace is substantially faster.

Related Reads

a16z: AI's 'Amnesia', Can Continuous Learning Cure It?

The article "a16z: AI's 'Amnesia' – Can Continual Learning Cure It?" explores the limitations of current large language models (LLMs), which, like the protagonist in the film *Memento*, are trapped in a perpetual present—unable to form new memories after training. While methods like in-context learning (ICL), retrieval-augmented generation (RAG), and external scaffolding (e.g., chat history, prompts) provide temporary solutions, they fail to enable true internalization of new knowledge. The authors argue that compression—the core of learning during training—is halted at deployment, preventing models from generalizing, discovering novel solutions (e.g., mathematical proofs), or handling adversarial scenarios. The piece introduces *continual learning* as a critical research direction to address this, categorizing approaches into three paths: 1. **Context**: Scaling external memory via longer context windows, multi-agent systems, and smarter retrieval. 2. **Modules**: Using pluggable adapters or external memory layers for specialization without full retraining. 3. **Weights**: Enabling parameter updates through sparse training, test-time training, meta-learning, distillation, and reinforcement learning from feedback. Challenges include catastrophic forgetting, safety risks, and auditability, but overcoming these could unlock models that learn iteratively from experience. The conclusion emphasizes that while context-based methods are effective, true breakthroughs require models to compress new information into weights post-deployment, moving from mere retrieval to genuine learning.

marsbit2h ago

a16z: AI's 'Amnesia', Can Continuous Learning Cure It?

marsbit2h ago

Can a Hair Dryer Earn $34,000? Deciphering the Reflexivity Paradox in Prediction Markets

An individual manipulated a weather sensor at Paris Charles de Gaulle Airport with a portable heat source, causing a Polymarket weather market to settle at 22°C and earning $34,000. This incident highlights a fundamental issue in prediction markets: when a market aims to reflect reality, it also incentivizes participants to influence that reality. Prediction markets operate on two layers: platform rules (what outcome counts as a win) and data sources (what actually happened). While most focus on rules, the real vulnerability lies in the data source. If reality is recorded through a specific source, influencing that source directly affects market settlement. The article categorizes markets by their vulnerability: 1. **Single-point physical data sources** (e.g., weather stations): Easily manipulated through physical interference. 2. **Insider information markets** (e.g., MrBeast video details): Insiders like team members use non-public information to trade. Kalshi fined a剪辑师 $20,000 for insider trading. 3. **Actor-manipulated markets** (e.g., Andrew Tate’s tweet counts): The subject of the market can control the outcome. Evidence suggests Tate’sociated accounts coordinated to profit. 4. **Individual-action markets** (e.g., WNBA disruptions): A single person can execute an event to profit from their pre-placed bets. Kalshi and Polymarket handle these issues differently. Kalshi enforces strict KYC, publicly penalizes insider trading, and reports to regulators. Polymarket, with its anonymous wallet-based system, has historically been more permissive, arguing that insider information improves market accuracy. However, it cooperated with authorities in the "Van Dyke case," where a user traded on classified government information. The core paradox is reflexivity: prediction markets are designed to discover truth, but their financial incentives can distort reality. The more valuable a prediction becomes, the more likely participants are to influence the event itself. The market ceases to be a mirror of reality and instead shapes it.

marsbit3h ago

Can a Hair Dryer Earn $34,000? Deciphering the Reflexivity Paradox in Prediction Markets

marsbit3h ago

Trading

Spot
Futures

Hot Articles

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of AI (AI) are presented below.

活动图片