From an investment of 13 billion US dollars to an expected return of 92 billion US dollars! Microsoft predicted and made a real money bet on “AI will set off a productivity revolution” 3 years ago

The Zhitong Finance App learned that the US tech giant Microsoft (MSFT.US) has set a return target of up to 92 billion US dollars for its large-scale early investment in ChatGPT developer OpenAI. This landmark investment arrangement can be said to have helped start the current AI application era. The impressive return on investment target of up to 92 billion US dollars shows that AI application platforms such as ChatGPT are not only a major breakthrough in the AI technology paradigm, but also a major path that can transform productivity and transform into huge cash flow through a series of market-popular commercial AI applications. Up to now, Microsoft's cumulative investment in OpenAI is about 13 billion US dollars. From the perspective of return on investment (ROI), this return is expected to be extremely high, highlighting the extremely strong prospects for AI applications to generate revenue.

This goal was included in Microsoft's internal planning document in early 2023 and publicly disclosed in a US court on Monday EST. Microsoft CEO Satya Nadella testified to the jury in a high-profile lawsuit against OpenAI and Microsoft by Elon Musk, the world's richest man and Tesla CEO, in federal court in Oakland, California, that the investment “actually progressed very well and had great benefits because we took the risk.”

As the world's largest software manufacturer, Microsoft previously invested about $13 billion in the ChatGPT maker through a round of financing in early 2023. Since then, OpenAI's valuation has continued to soar, and the public market valuation as of the end of March had reached $852 billion. According to this internal planning document released in court, as of October last year, Microsoft's shares in the company were valued at about 135 billion US dollars.

In a lawsuit in 2024, Tesla CEO Musk accused OpenAI co-founders Sam Altman and Greg Brockman of completely betraying the AI startup's major founding mission as a charitable non-profit organization to benefit humanity when switching to a for-profit operating model. The world's richest man launched this high-profile lawsuit after claiming that Microsoft also helped with this betrayal.

Microsoft and OpenAI have had disputes over terms of cooperation, and over time, the two have gradually formed a more direct competitive relationship for AI applications. As an important part of OpenAI's major restructuring last year, Microsoft acquired approximately 27% of the AI startup's ownership. From the perspective of return on investment (ROI), Microsoft's return expectations are extremely high. Microsoft's $13 billion investment was exchanged for shares worth about $228 billion (based on a valuation of US$852 billion and a shareholding ratio of about 27%), representing a book return close to 17.6 times, yet this return does not include OpenAI's strategic technology embedding and the hundreds of billion dollars of Azure cloud AI computing power resources OpenAI promised to buy in the future.

OpenAI, Altman, Brockman, and Microsoft management all denied any improper operation, saying Musk's accusations were unfounded harassment aimed at increasing the actual valuation of xAI, an AI startup launched by itself in 2023.

A New Era of AI Commercialization: Anthropic and OpenAI Reshape Software Companies' Revenue Generation Blueprint

After attending the HumanX Global Artificial Intelligence Conference in San Francisco, analysts from international investment giant UBS (UBS) recently said that the world's two most cutting-edge AI application leaders — OpenAI and Anthropic — are transforming from AI supergiants to “devours” of corporate IT budgets. UBS analysts write that the world's most advanced AI companies, such as Anthropic and OpenAI, have introduced AI models with stronger comprehensive capabilities, which indeed poses a major threat to the fundamental growth prospects of most traditional software companies — these cutting-edge AI companies already have strong enough ability to seize “wallet shares” from larger enterprise customers.

Enterprise wallet share mainly refers to how much a technology industry vendor has taken away from the total IT/software/digital budget of enterprise customers. According to the UBS analyst team, the enterprise-level IT budgets that enterprise customers would otherwise spend on a large scale on traditional software vendors are being taken away by cutting-edge AI companies such as OpenAI and Anthropic.

Microsoft's early large investment in OpenAI and the expected return target of 92 billion US dollars set internally, combined with Anthropic's explosive revenue growth trajectory, can be described as an extremely optimistic commercialization path for AI agents and cutting-edge enterprise-level AI application products. Anthropic announced last month that the company's annualized revenue data (ARR) had exceeded 30 billion US dollars, a significant increase compared to 9 billion US dollars at the end of 2025; while the third-party research agency Semi Analysis pointed out in a report in early May that Anthropic's annualized revenue had risen to about 44 billion US dollars. This growth rate far exceeds that of OpenAI over the same period.

The surge in demand for Claude has become Anthropic's “sweet burden.” Anthropic CEO Amoudi said last week that the company was trying to plan for 10-fold growth, but revenue and usage increased 80 times on an annualized basis in the first quarter, which also explains why it is difficult for the company to meet demand. He also stated slightly “Versailles” that the current level of growth is “just crazy” and “difficult to handle”, and he hopes future expansion will be “more normal.”

Nadella emphasized that “the investment results are good because we took risks”. This statement essentially reflects the adoption of cutting-edge AI technology — especially enterprise-level APIs, AI smart assistants that focus on agent-based workflows, and generative AI applications — have strong monetization capabilities and can redefine the software industry's revenue structure (from one-time commercial licensing to ongoing subscriptions and API billing). This investment case itself provides a strong verification framework for the revenue generation path of AI applications, that is, from technological explosion to large-scale enterprise adoption, to a closed commercial loop that stabilizes the cash flow path.

More than 80% of Anthropic's revenue comes from enterprise customers, of which more than 1,000 companies spend more than $1 million a year on the Claude AI platform. The AI application leader's growth relies not only on Claude's core big language model, but also on Claude Code, Claude Cowork, and similar pre-built AI agents (AI agents) for vertical industries (such as financial services). These tools can replace traditional automation and analysis within a few days of deployment, greatly improving work efficiency and business output.

US productivity data strengthens AI investment logic: enterprises increase intelligence, AI application bull market is far from over

AI agent economics is becoming the standard model for future commercial software: OpenAI/Anthropic's deployment path clearly reflects a trend: enterprise customers are more willing to pay expensive contracts for AI agents that can automate complex tasks, integrate industry data sources, and meet regulatory requirements, rather than just API usage practices. This combination of agents (skills+connectors+sub-models) creates “subscriptable, customizable, and expandable” proxy AI workflows in high-value fields such as finance, law, and technology development, significantly increasing stickiness and life cycle value. In addition, training techniques introduced by Anthropic, such as “AI dreaming,” enhance the ability of its agents to learn independently, which will further promote the evolution of intelligent entities from passive tools to semi-autonomous business assistants, and significantly enhance the marginal value of commercialization.

According to the latest economic data, US labor productivity continued to rise in the first quarter. Although the growth rate slowed slightly, this growth trend shows that companies are using cutting-edge AI technology to gradually improve employee efficiency to ease the pressure on higher energy costs brought about by the geopolitical conflict in the Middle East. A number of recent productivity data reports prove that companies are treating cutting-edge AI technologies such as AI agents that focus on agent-based workflows as “cost reduction and efficiency capital expenditure,” strengthening the judgment that “enterprise AI investment is still early, and AI application demand is still in the long run.” It also supports the continued sharp rise in Anthropic/OpenAI valuations and the long-term stock price trajectory of cloud computing supergiants such as Microsoft and Google.

According to data released by the US Bureau of Labor Statistics on Thursday, the productivity indicator, the hourly output of non-farm workers, grew at an annualized quarterly rate of 0.8% after achieving 1.6% growth in the fourth quarter after being lowered and revised. The more impressive data is that under the year-on-year standard, productivity rose significantly by 2.9% compared to a year ago, the biggest annual increase since 2024, highlighting the strong productivity expansion brought about by investment in AI smart devices. In fact, the US Bureau of Labor Statistics said that the share of output received by workers in the form of pay in the first quarter was only 54.1%, the lowest recorded value since the beginning of this series of data in 1947. The US Bureau of Labor Statistics report also showed that manufacturing productivity recorded its biggest increase in a year in the first quarter, rebounding significantly from the fall at the end of the year.

The recent trend of increasing labor productivity helps ensure that wage pressure is no longer the main source of rising inflation, which confirms the latest views of Federal Reserve officials. Companies have increased their spending on emerging technologies such as artificial intelligence in recent years to help ease pressure from other rising costs, such as surging costs associated with tariffs or the Iran war. Increased productivity also positively indicates that under the pressure of tariffs, oil prices, and wages, enterprises are indeed more willing to increase per capita output through AI, automation, and software tools than simply expand recruitment.

From the perspective of AI engineering and enterprise IT, the transmission chain logic of AI application leaders such as Microsoft, Google, and Oracle can be described as continuing to strengthen — enterprises pursue productivity improvement, then increase AI tool/automation/agent deployment, and ultimately drive demand for cloud AI computing power resources, model APIs, enterprise AI application kits, code agents, and office Copilot continues to rise. Therefore, the latest enterprise productivity data can be described as strengthening the judgment that “enterprise AI investment is still early, and AI application demand is still falling and snowy.”

The urgent need for companies to improve efficiency and reduce operating costs can be described as vigorously promoting the two core categories of AI application software — generative AI applications and the widespread application of AI agents. Among them, AI agents (AI agents, AI agents) that independently perform various complicated and complicated tasks will most likely be the ultimate trend in AI applications over the next 10 years. The emergence of AI agents means that artificial intelligence is beginning to evolve from an information aid tool to a highly intelligent productivity tool. This is why Anthropic's valuation can exceed 1 trillion US dollars and surpass OpenAI.