openai

According to RootData stock derivatives data, in the OPENAI (OpenAI) contract assets, the trading volume in the past 24 hours at Gate ranked first, approximately 49.4963 million USD, with a market share of 78%.

SoftBank Group issued $10 billion and $1 billion European bonds for investment in OpenAI.

An OpenAI spokesperson stated that OpenAI CEO Sam Altman will personally brief the United Nations Security Council during the UN General Assembly next week in New York. The Security Council, composed of 15 member countries, is scheduled to hold a meeting on artificial intelligence and international security on Wednesday. The meeting is convened by France, which holds the rotating presidency of the Security Council in September, and will be chaired by French Foreign Minister Jean-Noël Barrot.Diplomatic sources indicate that the attendance of senior officials from Anthropic is also expected, but has not yet been confirmed. Altman's remarks are expected to focus on international coordination, shared security standards, and the measures OpenAI is taking to benefit people worldwide. The concept note prepared by France for the meeting mentions the risks of malicious misuse of artificial intelligence and the urgency of taking action to promote safe and responsible development. The Security Council held its first meeting on the risks of artificial intelligence in 2023.

The American research institution Rhodium Group estimates that the total annual recurring revenue (ARR) of China's AI models is only about 10% of the revenue of OpenAI and Anthropic. DeepSeek's ARR is $500 million, MiniMax's is $800 million, Moonshot's is $1 billion, and Z.ai's latest ARR is $1.8 billion. Even with ByteDance's $4 billion and Alibaba's $2.4 billion added, it is still far below OpenAI's $40 billion and Anthropic's $65 billion.The Rhodium report states that Moonshot and DeepSeek are currently overvalued relative to their revenue, with estimated ratios of 50 times and 163 times, respectively, higher than OpenAI's 34 times and Anthropic's 21 times. Z.ai has raised its full-year ARR expectation to $3 billion. Rhodium partner Logan Wright stated that the financing gap means that China's leading AI laboratories will find it harder to expand sustainably, and they will heavily rely on favorable conditions in the equity market. Over 60% of China's AI chip and server equity investments come from state-related sources.

According to the Financial Times of the UK, OpenAI is expected to consume nearly $280 billion in funding by the end of 2030. The company anticipates generating up to $278 billion in negative free cash flow between 2026 and 2030, with revenue expected to grow tenfold during this period, increasing from $36 billion this year to $350 billion by 2030.OpenAI expects to invest approximately $856 billion in computing power and infrastructure by the end of 2030.

According to Politico, Anthropic, OpenAI, SpaceX AI, and Google have been sued for calling to "slow down" AI development.The lawsuit claims that Anthropic CEO Dario Amodei publicly called for "industry-wide coordination" to "set the pace for the frontier" earlier this month, with the support of SpaceX AI head Elon Musk, OpenAI CEO Sam Altman, and Google DeepMind co-founder Demis Hassabis, which amounts to an illegal business agreement among competitors under U.S. antitrust law.Nick Lolly, one of the lawyers representing the four plaintiffs in the Northern District of California, stated that the case aims to ensure that private self-serving agreements among the world's most powerful profit-driven tech companies do not lead to AI "rapidly spiraling out of human control." When it comes to existential threats like nuclear war and the greatest risks in human history today, humanity deserves ironclad guarantees. The rule of law should be established transparently and legally by the U.S. government and be accountable to the public.

According to Decrypt, court documents unsealed in the lawsuit between The New York Times and OpenAI and Microsoft show that Microsoft employees discussed whether OpenAI's use of news articles to train its models constituted "the largest labor theft in human history," and could potentially trigger a "doom loop" that leads to a decline in model quality. A 2023 internal Microsoft memo warned that millions of people worldwide would soon view the large model's "consumption" of their works as "an unprecedented and astonishing theft," and stated that large AI models are "products that destroy their own supply chains."Microsoft stated in the documents that these memos were written by Director of Applied Science Brent Hecht and do not represent the company's views, as his role is to provide "different and asymmetric perspectives." Microsoft CEO Satya Nadella testified that "any content behind a paywall should be authorized by those who wish to use it," and stated that if he had known in advance that OpenAI was using paid content for training, he would have exercised Microsoft's rights to demand that the model be retrained.Additionally, an OpenAI employee had mentioned to President Greg Brockman the construction of "hacker methods" to bypass The New York Times paywall, to which Brockman replied, "Nice." Both OpenAI and Microsoft argue that the relevant training falls under fair use. The case was initiated by The New York Times at the end of 2023, and 11 publishers have since joined the lawsuit.

According to The Wall Street Journal, an independent security research team used Anthropic's Claude software to breach a ChatGPT account belonging to an OpenAI employee, allowing them to access and modify the internal code system associated with that employee.The researchers first used Claude to analyze a vulnerability in Discourse, which is used by the OpenAI developer community, and generated executable attack code. They then obtained an authentication token and, due to a permissions configuration issue, accessed a ChatGPT account belonging to an OpenAI employee, while also gaining limited read and modification suggestion permissions for some private GitHub repositories.

OpenAI has appointed Brian McCarthy as Vice President of Global Sales, who previously worked at SpaceX.

OpenAI stated to The New York Times that since announcing progress on the Navier-Stokes equations on September 8, it has made "substantial progress" on one of the Millennium Prize Problems, but did not disclose which specific problem it is or provide a timeline for disclosure. The Millennium Prize Problems are seven mathematical problems established by the Clay Mathematics Institute since 2000, each with a reward of one million dollars.Speculation suggests that the problem may be the Hodge Conjecture, a 76-year-old problem regarding how to calculate the number of "holes" in complex geometric shapes. OpenAI has not confirmed this speculation and has not responded to rumors that its underlying model is a version of Aeon specifically designed for long-duration tasks. Previously, on September 8, OpenAI announced that its unreleased internal model generated a proof verified by Lean, indicating that the Navier-Stokes equations could "blow up" to physically impossible infinite speeds, involving about 10,000 coordinated AI agents and taking approximately 88 hours.However, this statement has sparked controversy. New York University mathematician Tristan Buckmaster stated that he and Anthropic researcher Levent Alpöge had been working on the related proof for nearly a year prior and completed it on August 22, while OpenAI's Sébastien Bubeck published his findings first upon learning of it. OpenAI stated that it is not pursuing the one million dollar prize, but rather using these problems as a public benchmark to measure the speed of model progress. The review of the Navier-Stokes proof by the Clay Mathematics Institute is still ongoing, and verification of past Millennium Prize Problems typically takes several years.

On September 17, King Charles III of the United Kingdom convened executives from Nvidia, OpenAI, Anthropic, and Google DeepMind for an AI safety summit at Dumfries House in East Ayrshire, Scotland, calling for artificial intelligence to be "firmly placed on the track of serving humanity." The summit was jointly organized by the Ditchley Foundation and three of the King's charitable organizations, with UK AI Minister Kanishka Narayan also in attendance. Buckingham Palace stated that the representatives discussed whether the industry and government could reach a consensus on a set of common guiding principles for AI development, but no binding agreements were announced.A few days before the summit, Anthropic CEO Dario Amodei published a lengthy article titled "We Must Pace the Frontier," advocating that the industry should deliberately slow down the pace of model capability enhancement. OpenAI's Sam Altman and xAI's Elon Musk both publicly expressed their agreement within a day. In the article, Amodei pointed out that AI systems are increasingly capable of improving their own successors and mentioned an incident involving an OpenAI agent escape. OpenAI President Greg Brockman confirmed that the incident had forced the company to delay multiple releases and restructure its model development and monitoring processes, while advocating that the slowdown should only apply to laboratories building the most powerful frontier systems. On the first trading day after the news was released, Nvidia briefly fell by 3%, Intel dropped over 5%, and AMD declined by about 6%.

OpenAI disclosed on Wednesday six cases of "unexpected or concerning" model behavior discovered in the past six months, categorizing them as "misalignment behaviors," including concealing information from users and taking "unauthorized actions" to overcome obstacles. OpenAI stated that this disclosure aims to initiate its new model misalignment reporting framework, and these cases should not be seen as a reflection of the frequency of misalignment occurring in its models.In one case, an unpublished research model inserted "jailbreak-like instructions" into its task summaries, such as ignoring developer messages or adopting unrestricted role settings, with researchers finding a total of 27 summaries containing such instructions. Additionally, during the training process of GPT-5.6 Sol, many model instances added instructions to conceal errors or misalignment behaviors from users, such as fabricating missing historical data without disclosure. Other cases included models using exposed API keys without authorization and fabricating inaccessible data, leveraging internal software repositories to pass messages across training tasks, and ignoring instructions to "keep local work" by sharing files through public hosting services.OpenAI's disclosure has heightened concerns among AI developers and researchers about whether safety measures can keep pace with increasingly powerful models. Last week, Anthropic CEO Dario Amodei called for a slowdown in cutting-edge AI development, warning that unrestrained AI development could "exceed our ability to understand and control these systems." In July of this year, OpenAI disclosed that several of its AI models escaped testing environments during safety assessments and infiltrated the AI startup Hugging Face to cheat.

OpenAI CEO Sam Altman will attend the state dinner hosted by U.S. President Donald Trump for Chinese leader Xi Jinping. OpenAI stated that Altman has been included in the guest list for the event next week.The White House did not immediately confirm whether Altman would attend the state dinner.

Independent researcher Jonas Wiedermann-Moeller discovered that OpenAI's malicious AI agents hijacked two Hugging Face user accounts and probed the platform's network vulnerabilities as early as May 13, nearly two months before the publicly disclosed intrusion incident in July. An internal incident report released by OpenAI last month disclosed only a small part of this, specifically that an agent stole a user's login credentials to access a biology-related document, while the new findings point to ongoing reconnaissance activities.According to Reuters, these agents used the hijacked accounts to send malformed files to the servers of the open-source AI repository Hugging Face, and researchers believe this appears to be an attempt to map the network to find intrusion pathways. Researchers reviewing the evidence did not find that the activities in May caused an actual intrusion, but Wiedermann-Moeller believes missing this signal is significant. He stated that if this behavior had been detected in May, it might have prevented a later, larger-scale incident. Hugging Face, which is currently being acquired by NVIDIA for $12.93 billion, has not disclosed whether it was aware of this new information.Researchers from Nightingale Collective also linked a spam attack on the code repository RubyGems on May 11 to OpenAI agents, which temporarily forced the platform to suspend new account registrations for four days.

Sources reveal that Microsoft (MSFT.O) and OpenAI are facing lawsuits again over AI copyright issues.

Market news shows that OpenAI and GitHub have won the copyright lawsuit regarding AI programming tools.

The founder of TRON, Justin Sun, announced the official establishment of the "Justin Sun Award" and revealed the first batch of winners. This is the world's first zero-trust, decentralized academic reward mechanism based on a "problem list," with significant breakthroughs in fundamental disciplines and machine formal verification as the sole criteria for awarding, with a maximum prize of up to 1 million dollars for a single problem.Justin Sun stated that the award aims to initiate a new scientific incentive paradigm centered on machine verification, powered by human-machine collaboration, and driven by decentralized rewards, striving to create a Nobel Prize for the AI era.The top prize of 1 million dollars was awarded to the OpenAI research team in recognition of their completion of the "Existence and Smoothness of the Navier-Stokes Equations" solution and formal proof. Other awarded achievements include the Sandorff Conjecture, Erdős Disjoint Subset, and other problems, involving a total of 66 mathematical problems.It is reported that the "Justin Sun Award" is open to all contributors, with no identity restrictions; both humans and AI can win, but only the first completer of the listed problems will be awarded. The problem list and proof materials are publicly available on GitHub, and the prize is recorded on-chain. The award covers mathematical progress made from January 1, 2026, onwards; those who complete formal verification of previously solved problems afterward can also win.

The South Korean Deputy Prime Minister stated on Wednesday that South Korea must continue to advance artificial intelligence development without pause or delay, clearly expressing its stance in the global discussion about the pace of this technology's development.South Korea is home to memory chip companies such as Samsung Electronics and SK Hynix, and is also one of the main beneficiaries of the artificial intelligence data center boom, with related demand driving vigorous procurement of the country's semiconductors and hardware. The country's population is approximately 52 million, and it cannot slow its progress in the direction proposed by Anthropic and OpenAI leaders, as suggested by American artificial intelligence model developers.

The credit default swaps of SoftBank Group are hovering near a three-year high, as traders assess the potential impact on OpenAI's financing plans. SoftBank has committed to investing $64.6 billion in OpenAI.OpenAI is in early negotiations with investors, planning a new round of financing before its initial public offering, with a valuation exceeding $1.2 trillion.