Sam Altman萨姆·奥尔特曼
OpenAI
CEO
United StatesSam Altman, male, U.S. national, born April 1985, a native of Chicago, Illinois, U.S., started his first venture in 2005, attended university without completing a degree. He is the CEO of OpenAI.
- 2003–2005Studied computer science at Stanford University in the United States (did not graduate)
- 2005–2012Co-founded Loopt, a mobile social networking company, and served as its CEO
- 2012–2014Co-founded the venture capital fund Hydrazine Capital
- 2014–2019President of Y Combinator, a U.S. startup accelerator(During this period: co-founded OpenAI in 2015 and served as co-chairman of its board)
- 2019–presentCEO of OpenAI
Wikipedia · Sam AltmanCompany
Frontier model demand and AI infrastructure driver.
Public event locations
Aggregates only locations that have appeared in public reports; locations, order and dates may be uncertain. The line and numbers indicate an inferred route only — not real-time positioning or current whereabouts.
Event timeline
Anthropic blocks all public access to Claude Fable 5, Mythos 5 following US government order
The US government issued an unprecedented export control directive ordering Anthropic to suspend access to its top-tier Claude Fable 5 and Mythos 5 models for foreign nationals, citing national security. Anthropic responded by blocking public access globally, affecting all users including enterprise customers and employees. The models had been released only three days prior and were recently jailbroken by Pliny the Liberator. Anthropic calls it a misunderstanding and is working to restore access.
AI costs spike as subscriptions hit pricing wall — firms turn towards Chinese LLMs, open-source models to extend budget
As token costs for frontier AI models skyrocket, companies are seeking cheaper alternatives, turning to Chinese LLMs and open-source models. Subscription models are also eating into the profitability of startups like OpenAI and Anthropic, with utilization rates above 5.7% potentially leading to losses.
OpenAI says it's engaging 'constructively' with state AGs about concerns
The Wall Street Journal reported that a coalition of state attorneys general has opened an investigation into OpenAI. OpenAI responded that it is engaging constructively with the state AGs on the concerns, without disclosing specifics. The probe may involve consumer protection and data privacy issues.
Microsoft’s open-source SkillOpt automatically upgrades AI agent skills without touching model weights
Microsoft released SkillOpt, an open-source framework that optimizes AI agent skill files (.md). Traditionally, skill optimization requires manual instruction editing, which is slow and error-prone. SkillOpt treats skill documents as trainable objects, automatically exploring optimal instruction combinations via deep-learning-style optimization, boosting accuracy for models like GPT-5.5 without altering underlying weights.
Another parent has filed a wrongful death suit against OpenAI
Another parent is suing OpenAI, claiming its chatbot did not do enough to prevent their child's death by suicide. The lawsuit alleges that OpenAI's product was defective and failed to protect minors from harm. This is one of several lawsuits against OpenAI, highlighting AI safety and liability issues.
OpenAI could go from AI pioneer to AI's BlackBerry, says Forrester
Forrester Research warns that OpenAI, while courting investors and chasing enterprise customers, could transition from today's AI leader to tomorrow's cautionary tale, akin to BlackBerry's decline. The report highlights intensifying competition, technological iteration risks, and commercialization challenges that could lead to OpenAI's downfall if strategic adjustments are not made.
OpenAI's Osborne on Governments Adopting AI
OpenAI's Osborne spoke on governments adopting AI, discussing public sector applications, policy frameworks, and safety considerations. He emphasized the need for cautious deployment while ensuring technology benefits the public. This reflects OpenAI's ongoing engagement in government AI policy.
OpenAI considers drastic price cuts, anticipating war for users with Anthropic, WSJ reports
OpenAI is considering drastically reducing prices charged to users to win customers from rival Anthropic, according to the Wall Street Journal. This move signals intensifying competition in the AI model market as OpenAI seeks to gain market share through pricing advantages.
Visa adds payment network into ChatGPT, letting AI agents shop and pay for users
Visa announced the integration of its payment network into ChatGPT, enabling AI agents to shop and pay on behalf of users. This partnership allows ChatGPT to process payments directly within conversations, eliminating the need for users to leave the chat interface. Visa stated that this move aims to simplify payment processes and explore AI applications in fintech. The feature is currently in pilot phase and may expand to more scenarios in the future.
Surprise upset: GPT-5.5 beats Claude Fable 5 on brutal new Agents’ Last Exam benchmark
UC Berkeley's RDI and over 300 experts launched Agents’ Last Exam (ALE), a benchmark measuring AI's ability to execute economically valuable long-horizon workflows. OpenAI's GPT-5.5 topped the leaderboard with 24.0% pass rate, beating Anthropic's new Claude Fable 5 (22.0%). ALE aims to bridge academic benchmarks and real labor impact, showing top models still fundamentally fail.
Why Andrew Yang is building instead of waiting for Washington
Andrew Yang's 2020 presidential campaign warned that automation and AI would hollow out the labor market and concentrate wealth, with ideas like Universal Basic Income seen as fringe. Now, Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman, and Senator Bernie Sanders all echo similar views. As an entrepreneur, Yang chooses to build solutions through ventures rather than wait for Washington.
Researchers say they trained a foundation model from scratch for about $1,500
Researchers at Sapient developed HRM-Text, using a Hierarchical Recurrent Model architecture, training a 1B-parameter foundation model from scratch for about $1,500 with only instruction-response pairs. The model achieves competitive performance with much larger open models, dramatically reducing pretraining costs and enabling organizations with limited resources to train their own reasoning models.
OpenAI says Chinese propaganda is being deployed to foment dissent over tariffs, data centers
In a report published on Wednesday, OpenAI stated that Chinese propagandists have been using its flagship chat tool to stir opposition to Donald Trump's tariffs and intervene in U.S. debates over data centers and AI. The activities aim to create social discord and influence policy discussions.
OpenAI Says China-Linked Accounts Aim to Fuel US Data Center Pushback
OpenAI stated that some China-linked social media accounts are spreading disinformation to incite public opposition against large-scale data center projects in the US. These accounts claim data centers consume excessive energy and harm the environment, but OpenAI believes their real aim is to hinder US progress in AI infrastructure. The company has taken steps to identify and flag these accounts.
OpenAI Joins Race for Public Market Cash
OpenAI is exploring the possibility of raising funds through public markets to support its massive AI research and infrastructure development. This move signals that the private company may be considering an IPO or bond issuance to secure more capital to maintain its leading position.
Anthropic brings Mythos to the masses with Claude Fable 5, its most powerful generally available model ever
Anthropic today launched two new AI models: Claude Fable 5 and Claude Mythos 5. Fable 5, the version for most users and developers, surpasses all previous generally available Claude models with stronger performance in software engineering, knowledge work, vision, scientific research, and long-running tasks. Mythos 5 is a restricted upgrade for approved users, including cybersecurity partners and select biology researchers. Fable 5 wraps Mythos-class capabilities in new safeguards, routing high-risk requests to Claude Opus 4.8. Over 95% of Fable sessions run on its own responses, with no universal jailbreaks found after 1,000+ hours of testing. Fable 5 is available to the public today; Mythos 5 initially only to existing restricted users.
SpaceX, Anthropic, OpenAI Can Rewrite History for Megacap IPOs
The article notes that unlisted tech giants like SpaceX, Anthropic, and OpenAI could reshape capital markets through massive IPOs. Their high valuations could set records if they go public, but profitability and regulatory challenges remain. Markets expect them to become new benchmarks.
Researchers train open source AI search agent Harness-1, outperforming GPT-5.4 on relevant information recall
A joint research collaboration between UIUC, UC Berkeley, and Chroma released Harness-1, a 20-billion parameter open-source search agent built on OpenAI's gpt-oss-20B model. Harness-1 achieves 73% average on recalling relevant information from a curated dataset, outperforming GPT-5.4 (70.9%) and Tongyi DeepResearch 30B by 11.4 percentage points. The model and code are available under Apache 2.0 on Hugging Face. It also validates the Tinker distributed AI training API by Thinking Machines.
OpenAI Kicks Off IPO Process in Test of Investor Appetite for Top AI Labs
According to the Wall Street Journal, OpenAI has initiated its initial public offering (IPO) process to gauge investor appetite for top AI labs. This marks a significant step in OpenAI's transition from a non-profit to a commercial entity, potentially bringing new capital into the AI industry. The specific timeline and valuation of the IPO have not been disclosed.
Following Anthropic, OpenAI files confidentially for IPO
ChatGPT-maker OpenAI announced Monday in a blog post that it has confidentially filed for an initial public offering (IPO). The filing comes just over a week after its main rival Anthropic also filed to go public, intensifying the race between the two AI firms. OpenAI was last valued at $852 billion post-money.
The Agentic Reckoning: Enterprise AI organizations have a runtime problem, not a model problem — and most are building the wrong solution
In Q1 2026, VentureBeat's Pulse Research surfaced the 'Governance Mirage': the gap between governance org charts and actual control layers. 43% said a central team owned AI governance; 23% couldn't agree on who owned it; 31% named vendor opacity as the biggest obstacle. The new research finds the failure point is the runtime, not the model. AI agents on stateless infrastructure cannot survive production: container restarts erase context, token costs breach business cases, hallucinations compound into catastrophic failures. Most engineering teams spend more time managing plumbing than building intelligence.
Ask HN: Why won't you be replaced by AI?
A Hacker News discussion questioning the 'human in the loop' argument, suggesting that as AI general intelligence improves, all jobs could be replaced by LLM-driven agentic systems. Comments discuss AI progress but no specific companies or products.
Show HN: One API Key for 45 AI Models – Pay per Token, OpenAI Compatible
ModelHub API launches a service that provides access to 45 AI models including OpenAI, Anthropic, Google, etc. with a single API key. It uses pay-per-token pricing and is fully compatible with the OpenAI API format, allowing developers to switch seamlessly. The service aims to simplify multi-model integration and lower the barrier to entry.
Trump, Sanders and Altman discuss public ownership in AI
According to AP News, former U.S. President Donald Trump, Senator Bernie Sanders, and OpenAI CEO Sam Altman are discussing public ownership in artificial intelligence. This cross-party dialogue involves public ownership models for AI technology, potentially influencing future AI governance and policy directions. Details of the discussions have not been disclosed, but it indicates that AI governance has become a focal issue across the political spectrum.
OpenAI is still working on that ‘super app’
According to a senior OpenAI employee, 'Chat is dead,' implying the company is developing a super app that goes beyond traditional chat interfaces, potentially integrating multiple AI features into a one-stop platform.
OpenAI plans ChatGPT 'superapp' overhaul ahead of listing, FT reports
OpenAI is planning its biggest ChatGPT overhaul yet, aiming to turn it into a 'superapp' with coding tools and AI agents to boost revenue ahead of a planned share listing, the Financial Times reported.
OpenAI unveils Lockdown Mode to protect sensitive data from prompt injection attacks
OpenAI has introduced a new 'Lockdown Mode' feature designed to protect sensitive data from prompt injection attacks. The mode reduces the risk of sensitive information leakage by limiting the model's response to certain external instructions. However, OpenAI acknowledges that even with Lockdown Mode enabled, ChatGPT may still be vulnerable to prompt injections, with the goal being to reduce the likelihood of sensitive data being shared in the process.
Microsoft AI chief says company was “set free” from OpenAI to pursue superintelligence
Mustafa Suleyman, CEO of Microsoft AI, disclosed in an exclusive interview with VentureBeat that a contractual change with OpenAI about six months ago granted his division formal authority to pursue 'superintelligence' using Microsoft's own researchers, data pipelines, and custom silicon. This signals a strategic inflection point: Microsoft is not abandoning OpenAI but building alongside it, with its first in-house model family released the same day.
OpenAI rolls out a Lockdown Mode for extra protection against prompt injection attacks
OpenAI has announced a Lockdown Mode, a more robust security feature for users who need extra protection against prompt injection attacks. This move targets a small set of users with specific security needs, highlighting OpenAI's ongoing commitment to AI safety.
Microsoft's AI Futurist explains how he uses Copilot and the real-world problems enterprises are solving with agents
At Build 2026, Microsoft emphasized agents are rapidly moving into production in enterprise systems, announcing Microsoft IQ as a context layer across GitHub Copilot, Microsoft Foundry, and Copilot Studio, along with Work IQ APIs, Fabric IQ, Foundry IQ, and Web IQ. It also launched Scout, a personal work agent, and seven new MAI models including MAI-Thinking-1. Microsoft AI Futurist Marco Casalaina (former VP of Azure AI products) spoke with VentureBeat about agent strategy and model-choice philosophy.
Try the new console experience in Amazon Bedrock, optimized for Anthropic- and OpenAI-compatible APIs
Amazon Bedrock launches a new console experience, allowing users to browse and compare the latest AI models side by side, organize work into projects with streamlined evaluation workflows, and access project-aware documentation with auto-prefilled code snippets ready to copy and run.
S&P 500 rejects SpaceX, also blocking entry for OpenAI and Anthropic
The S&P 500 index has rejected SpaceX from inclusion, also blocking OpenAI and Anthropic. This means these companies cannot easily access billions of dollars from passive investment funds. The move reflects the S&P 500's strict standards for non-traditional corporate structures, especially for unprofitable or specially governed companies.
Trump administration, OpenAI discussing possible government stake in the AI startup
OpenAI CEO Sam Altman first shared the idea with the Trump administration in 2025, according to a source. The two sides are now discussing a possible government stake in OpenAI, marking an unprecedented partnership model between the government and a leading AI company.
Model routing is a fix for AI overspending. That's a problem for OpenAI and Anthropic
Companies are shifting from using the most powerful AI model for all tasks to matching each task to the right model, a practice called model routing. This helps reduce AI costs but may decrease demand for high-end model providers like OpenAI and Anthropic, whose models are typically more expensive.
LinkedIn co-founder Reid Hoffman is leaving Microsoft's board after almost a decade
LinkedIn co-founder Reid Hoffman is leaving Microsoft's board after almost a decade. Hoffman was one of OpenAI's first donors, and he left its board in 2023 as the nonprofit cozied up with Microsoft.
AI is designing OpenAI's next model in a sign of 'super intelligence': SoftBank's Masayoshi Son to CNBC
SoftBank CEO Masayoshi Son told CNBC that his previous forecast of artificial super intelligence arriving in ten years was 'conservative' and believes it will happen sooner. He revealed that AI is already designing OpenAI's next model, signaling super intelligence. Son emphasized that AI development is outpacing expectations and SoftBank will continue to invest heavily in AI.
OpenAI will let the US government review its AI models before release
OpenAI said this week it would comply with President Trump's voluntary AI executive order, allowing the US government to review its AI models before public release. This move aims to ensure AI safety and align with national interests, reflecting cooperation between the company and the government on AI regulation.
Market reaction
The following is market reference information for related companies, and does not constitute investment advice.