openai

SoftBank Group is seeking a record loan of up to $40 billion to acquire a stake in OpenAI.

OpenAI announced the release of GPT-5.4 in ChatGPT (in the form of GPT-5.4 Thinking), API, and Codex, marking its most powerful and efficient state-of-the-art model for professional work to date. It also released GPT-5.4 Pro for ChatGPT and API users, designed specifically for those who seek extreme performance on complex tasks.GPT-5.4 integrates OpenAI's recent advancements in reasoning, programming, and agent-based workflows into a single state-of-the-art model, incorporating the industry-leading programming capabilities of GPT-5.3-Codex while improving the model's performance in tools, software environments, and professional tasks involving spreadsheets, presentations, and documents.

According to TheInformation, as of the end of February 2026, OpenAI's annualized revenue has exceeded $25 billion.OpenAI's Chief Financial Officer Sarah Friar confirmed in the official blog and multiple reports in January 2026 that the company's annualized revenue for 2025 surpassed $20 billion, representing an approximately 230% increase from $6 billion in 2024. However, some subsequent public investor information indicates that OpenAI's actual total revenue for 2025 was about $13.1 billion, exceeding the company's earlier target of $10 billion but falling short of the peak annualized estimate of $20 billion.After rapid growth in 2025, OpenAI experienced a slowdown in growth in the second half of the year. Media outlets like CNBC estimate that its mid-year annualized revenue last year was around $10 to $13 billion, with year-end annualized revenue at about $5.5 billion.In contrast, OpenAI's direct competitor Anthropic had a mid-year annualized revenue of about $4 billion in 2025, which quickly grew to around $9 billion by the end of the year, thanks to the surge in enterprise customers and products like Claude Code. As of early March this year, the latest annualized revenue has exceeded $19 billion. Currently, Claude Code's run rate has surpassed $2.5 billion, with over 500 clients spending more than $1 million annually. The gap between Anthropic and OpenAI is narrowing.

According to market news, NVIDIA CEO Jensen Huang stated at the Morgan Stanley Technology, Media & Telecom Conference that NVIDIA's recent $30 billion investment in OpenAI may be the last investment. The reason is that OpenAI plans to go public, possibly by the end of this year. Huang also mentioned, "I think the opportunity to invest $100 billion in OpenAI may not come again." So this may be our last chance to invest in a company of such significant importance.NVIDIA's $10 billion investment in OpenAI competitor Anthropic may also be the last one. Huang reiterated that the deployment of AI computing power is already bringing substantial profit revenue to enterprises, including large publicly traded data center operators like Microsoft. He believes that if these clients can obtain more computing power, their growth rate will accelerate even faster; with a threefold increase in computing power, their revenue is also expected to triple. He also stated that automation design software and system suppliers like Cadence and Synopsys will become larger and more important in the future.

According to sources, the former Chief Research Scientist of OpenAI is raising $70 million for an artificial intelligence startup that is developing software to help factories achieve autonomous operation.

After the Trump administration in the United States discontinued the use of Anthropic AI technology products, OpenAI quickly reached an agreement with the U.S. Department of Defense to deploy OpenAI models within the Department's secure networks. This event triggered a backlash from OpenAI users, leading to a boycott movement, with 700,000 people canceling their subscriptions and even uninstalling ChatGPT, turning instead to Anthropic's Claude.With significant user support, Anthropic's Claude topped the free download charts in the U.S. app store, while OpenAI's ChatGPT and Google's Gemini ranked second and third, respectively. Claude had previously experienced an increase in failure rates, which may be related to the significant rise in traffic.

Anthropic is accused of launching a prompt tool for "exporting ChatGPT memory data," helping users transfer historical memory information to its model Claude, which has drawn industry attention.According to public content, the tool allows users to export their memory data from OpenAI's ChatGPT by copying and pasting specific prompts, and then import it into Claude. Related discussions suggest that this move is seen as directly weakening the user stickiness and switching costs that ChatGPT relies on through its "memory function."Market views consider the memory mechanism as an important moat for large model products— the longer users engage, the deeper the model understands their preferences, context, and historical conversations, thereby increasing the migration costs. If third-party tools can facilitate easy data migration, it may change the current user lock-in logic of AI products.Meanwhile, reports also mentioned that Anthropic was previously restricted from use by relevant systems of the U.S. Department of Defense, but the company's popularity and attention have quickly surged, topping some application rankings.Currently, the specific compliance of the aforementioned tool and the platform's response have not been clarified. The industry generally believes that competition among large models has extended from performance comparisons to ecological and data sovereignty aspects, with user data portability potentially becoming a key variable in the next stage.

OpenAI founder Sam Altman released a content summary after the latest AMA. In response to what would happen if the U.S. government attempted to nationalize OpenAI or other AI projects, Sam Altman stated that he could not predict the outcome. While it may be more appropriate for the government to lead the development of AGI in the long term, nationalization does not seem highly likely given the current trajectory of development.However, he emphasized that establishing a close collaborative relationship between the government and AI companies is "extremely important." People often take the "assurances" regarding safety for granted, but they tend to underestimate the significant investment and effort required behind them. He believes this is generally a good thing, but more respect and understanding are still needed.

According to Jinshi Data, sources say that Nvidia is preparing to launch a dedicated chip to shake up the competitive landscape of artificial intelligence computing. OpenAI is set to become one of the largest customers of Nvidia's new processor.

According to Fortune, OpenAI founder Sam Altman stated that OpenAI is in discussions with the U.S. Department of Defense to reach a potential agreement on the use of its AI models and tools, although the contract has not yet been signed. OpenAI will retain control over the implementation of technical safeguards, the deployment targets of the models, and the regions, limiting the deployment scope to cloud environments rather than "edge systems." In a military context, edge systems may include devices such as aircraft and drones.As a significant concession, the U.S. government has agreed to incorporate the "red lines" set by OpenAI into the contract terms, including a ban on using AI for autonomous weapon systems, domestic mass surveillance, and critical decision-making areas.

According to WIRED, OpenAI recently fired an employee for trading on prediction market platforms like Polymarket and Kalshi based on insider information.The report states that the employee used non-public information obtained through their work at OpenAI to place bets on related contracts, which is suspected to constitute "insider trading" and violate the company's compliance policies.

According to Gate's market data, U.S. stocks opened with the Dow Jones Industrial Average down 0.75%, the S&P 500 down 0.82%, and the Nasdaq Composite down 1.2%. Netflix (NFLX.O) opened up 10.83% after the company announced its withdrawal from the acquisition bid for Warner Bros. Discovery (WBD.O). Amazon (AMZN.O) fell 0.44% and announced a partnership with OpenAI, investing $50 billion in OpenAI's new round of $110 billion investment.

According to official news, OpenAI announced a new investment of $110 billion at a valuation of $730 billion.This round of financing includes $30 billion from SoftBank, $30 billion from NVIDIA, and $50 billion from Amazon, with more financial investors expected to join in the future. This round of financing increases the value of the OpenAI Foundation's stake in the OpenAI Group to over $180 billion.Additionally, OpenAI also announced a multi-year strategic partnership with Amazon and expanded its collaboration with NVIDIA, which will provide 3GW of dedicated inference capacity and 2GW of training capacity using the Vera Rubin system.Currently, the weekly active user count for ChatGPT has surpassed 900 million, with over 50 million individual subscribers. The weekly active user count for Codex has doubled since the beginning of the year to 1.6 million, and the number of paid enterprise users exceeds 9 million.

According to official news from OpenAI, OpenAI and Amazon have announced a multi-year strategic partnership. Amazon will invest $50 billion in OpenAI, with an initial investment of $15 billion, and the remaining $35 billion will be completed in the coming months upon meeting specific conditions.The two parties will jointly develop a Stateful Runtime Environment powered by OpenAI models, which will be made available to AWS customers through Amazon Bedrock, expected to launch in the coming months. AWS will become the exclusive third-party cloud distribution provider for the OpenAI Frontier platform, which will be available for enterprises to build, deploy, and manage AI agent teams.In terms of infrastructure, the two parties will expand their existing $38 billion multi-year agreement to $100 billion, with a term of 8 years. OpenAI commits to using approximately 2 gigawatts of Trainium computing power through AWS infrastructure, covering Trainium3 and the next-generation Trainium4 chips expected to be delivered starting in 2027. Additionally, OpenAI and Amazon will collaborate to develop custom models for Amazon developers for consumer-facing applications, complementing Amazon's existing Nova series models.

According to the Financial Times, Nvidia is set to complete a $30 billion investment in OpenAI, replacing the $100 billion long-term investment commitment the two companies reached last year, as part of OpenAI's new round of large-scale financing.Sources reveal that negotiations have entered the final stages, with the investment potentially being completed as early as this weekend. OpenAI's new round of financing is expected to raise over $100 billion, bringing its valuation to $730 billion, not including the new funds. It is reported that OpenAI will reinvest most of the incremental funds into purchasing Nvidia hardware.

According to Bloomberg technology reporter Ed Ludlow, citing multiple insiders, OpenAI is close to finalizing the first phase of a new round of financing, with an expected amount exceeding $100 billion.The overall valuation (including this financing) could surpass $850 billion, with the initial phase of funding primarily coming from strategic investors, including giants like Amazon, SoftBank, Nvidia, and Microsoft.

According to The Block, OpenAI has announced a partnership with the crypto investment firm Paradigm to launch EVMbench, aimed at assessing the capabilities of AI Agents in the field of smart contract security, including the identification, patching, and exploitation of high-risk vulnerabilities.Both parties stated that as AI improves its capabilities in code writing and execution, it can be exploited by attackers but can also serve as a defensive tool. Therefore, it is necessary to test the security capabilities of AI systems in environments with real economic significance. EVMbench will provide various testing modes, including vulnerability detection, contract modification, elimination of exploitability, and simulating complete attack processes in a sandbox chain environment.At the launch of this tool, the DeFi protocol Moonwell and the cross-chain liquidity protocol CrossCurve have recently encountered smart contract vulnerability attacks, raising concerns about AI-assisted code security. EVMbench is reportedly built on 120 selected vulnerability samples, covering multiple public audits and competition cases.Industry analysts believe that as AI technology advances, the threshold for identifying smart contract vulnerabilities may further decrease, and the game between security defense and attack capabilities is accelerating.