OpenAI

OpenAI is an American artificial intelligence research organization founded in December 2015 and headquartered in San Francisco. It develops large-language models, image and video generation systems, and the consumer product ChatGPT. As of April 2026, it is among the leading frontier AI labs by capability, valuation, and consumer reach.

At a glance

• Founded: 2015 in San Francisco
• Paraform Talent Density: Rank #2, score 0.805 (May 2026). Source.
• Status: Private. Capped-profit subsidiary OpenAI Global LLC under nonprofit OpenAI Inc.
• Funding: Approximately $180 billion cumulative across 12 disclosed rounds. Most recent: a $122 billion round closed March 2026 at an $852 billion post-money valuation, with Amazon (up to $50B), Nvidia ($30B), and SoftBank ($30B) as the largest participants.
• CEO: Sam Altman
• Other notable leadership: Greg Brockman (President), Jakub Pachocki (Chief Scientist), Mark Chen (Chief Research Officer)
• Open weights: Mostly closed. The gpt-oss MoE family (August 2025) and Whisper are open weights. Other models including GPT, o-series, Sora, and DALL-E are closed.
• Flagship models: GPT-5.5 (April 2026), o-series reasoning models, Sora (video), DALL-E 3 (image)

Origins

OpenAI was founded in December 2015 as a nonprofit research organization. Founders included Sam Altman, Greg Brockman, Ilya Sutskever, and Elon Musk, with a stated mission of advancing artificial general intelligence (AGI) safely. In 2019 the organization restructured into a capped-profit entity, OpenAI Global LLC, under the original nonprofit OpenAI Inc., enabling Microsoft's $1 billion investment and the long-term Azure compute partnership.

ChatGPT launched in November 2022 and reached 100 million monthly active users within two months, the fastest consumer software adoption recorded at the time. In November 2023, the nonprofit board removed Sam Altman as CEO, then reinstated him within five days following a near-total employee revolt and Microsoft pressure. The episode prompted a partial board restructuring, after which Altman returned with reduced board independence. The capped-profit and nonprofit structures have remained, though further reorganization has been reported.

Mission and strategy

OpenAI's stated mission is to ensure that artificial general intelligence benefits all of humanity. The company describes AGI as "a system that can solve human-level problems" and frames its research roadmap as a path toward that capability.

The operational strategy combines three elements: frontier model scaling, consumer product distribution through ChatGPT, and large-scale infrastructure investment through the Stargate joint venture with Microsoft, Oracle, and SoftBank. CEO Sam Altman summarized the approach at an editors' lunch in spring 2026: "make the best models, build the best product around it and have enough infrastructure to serve it at scale."

The competitive premise underlying the strategy is that the lab that reaches AGI first will define the subsequent AI ecosystem. This has shaped OpenAI's release cadence (GPT-5.5 shipped six weeks after GPT-5.4), capital deployment (the Stargate initiative has a publicly stated $500 billion compute target), and product mix (broad consumer reach plus the 2026 enterprise expansion).

Models and products

• GPT-5.5. April 2026 flagship of the GPT family, released six weeks after GPT-5.4 (March 2026) and following GPT-5 (August 2025). All closed weights, available through the API and ChatGPT.
• o-series reasoning models. o1 (September 2024) and o3 (early 2025) introduced reinforcement learning on chain-of-thought reasoning. The reasoning capabilities are integrated into GPT-5.5 Pro variants.
• Sora 2. Video generation. Sora launched in December 2024 after a February 2024 research preview; Sora 2 followed in September 2025. Competes with Runway, Kuaishou's Kling, and Google's Veo.
• DALL-E 3. Image generation, November 2023. Bundled into ChatGPT for distribution; trails Black Forest Labs and Midjourney on creative output benchmarks.
• Whisper. Audio transcription, open weights. Released in 2022 and updated subsequently.
• gpt-oss. Open-weights MoE models in 20B and 120B parameter variants (gpt-oss-20b at 21B total / 3.6B active; gpt-oss-120b at 117B total / 5.1B active), released in August 2025. Positioned by OpenAI as a contribution to the open-source ecosystem.

OpenAI's product surface includes ChatGPT (Free, Plus, Pro, Business, and Enterprise tiers), the developer API, the Custom GPTs marketplace, Codex (a coding agent integrated into GPT-5.5), and Operator (a computer-use agent in beta).

Benchmarks and standing

As of April 2026, GPT-5.5 leads most major frontier benchmarks. Artificial Analysis Intelligence Index reports GPT-5.5 at 60.24, with Claude Opus 4.7 at 57.28. LMArena ELO scores place GPT-5.5 at 1328 (general), 1456 (coding), and 1389 (vision). GPT-5.5 also leads on GPQA Diamond (94.2%), ARC-AGI Challenge (92.3), and AIME 2025 (96.7%). The exception is SWE-bench Verified, where Claude Opus 4.7 leads at 87.6% against GPT-5.5 at 68.5%.

Frontier benchmark leadership shifts with each major model release across labs, so these positions are point-in-time.

Leadership

As of April 2026, OpenAI's senior leadership includes:

• Sam Altman, Chief Executive Officer. Reinstated in November 2023 after a five-day firing by the nonprofit board. Public voice for the company on capital strategy, IPO timeline, and AGI framing.
• Greg Brockman, President and co-founder. Longest-tenured technical leader. Briefly resigned during the November 2023 board crisis and returned with Altman.
• Jakub Pachocki, Chief Scientist. Took the role in 2024 following Ilya Sutskever's departure.
• Mark Chen, Chief Research Officer. Took the role in September 2024 following Bob McGrew's departure.

A series of senior departures in 2024 and 2025 reshaped the broader Frontier landscape. Mira Murati (former CTO) founded Thinking Machines Lab. Ilya Sutskever (former Chief Scientist) founded Safe Superintelligence. John Schulman (co-founder) briefly joined Anthropic in August 2024 before moving to Thinking Machines Lab as Chief Scientist in February 2025. Barret Zoph (former VP of Research, Post-Training) joined Thinking Machines as Chief Technology Officer alongside Murati and returned to OpenAI under Fidji Simo in January 2026. Zoph's successor as VP of Research for Post-Training, Liam Fedus, departed in March 2025 to co-found Periodic Labs. Bob McGrew (former CRO) is associated with a stealth project. Two of these new ventures, Thinking Machines and Safe Superintelligence, are now classified as Insurgent labs in their own right.

Funding and backers

OpenAI has raised approximately $180 billion across 12 disclosed rounds. The funding history includes Microsoft's strategic commitments totaling $13 billion across three tranches (a $1 billion investment in July 2019, an approximately $2 billion follow-on in 2021, and a $10 billion commitment in January 2023), plus a separate October 2025 commitment of $250 billion in Azure compute spending over multiple years.

Beyond Microsoft, the disclosed rounds include an October 2024 Series G of $6.6 billion at a $157 billion post-money valuation, a November 2024 SoftBank tender offer of approximately $1.5 billion in employee shares, a $40 billion primary round in March 2025 at a $300 billion valuation, an October 2025 employee tender offer of approximately $6.6 billion at a $500 billion valuation (with up to $10.3 billion authorized but not fully sold), and most recently a $122 billion round closed in March 2026 at an $852 billion post-money valuation. The March 2026 round was led by SoftBank with Amazon committing up to $50 billion, Nvidia $30 billion, and SoftBank $30 billion, plus approximately $12 billion from a broader investor pool and $3 billion from individual investors.

The capital structure is unusual. The nonprofit OpenAI Inc. holds governance over the commercial subsidiary OpenAI Global LLC. Restructuring to remove the nonprofit cap has been reported repeatedly through 2025 and 2026, and an IPO timeline has been discussed publicly without formal commitment.

Burn rate has been the focus of public reporting on the company's financial trajectory. The Information's coverage of leaked OpenAI projections in late 2025 placed full-year 2025 cash burn at approximately $9 billion against revenue of $13 billion in ARR, and projected 2026 burn at $17 billion against revenue of approximately $30 billion in ARR. OpenAI does not turn cash-flow positive in those projections until 2030, with cumulative losses through 2029 projected at approximately $115 billion. The March 2026 round added approximately $122 billion in committed capital against that burn profile.

Beyond Microsoft, the largest disclosed enterprise contract is the Cristal Intelligence joint venture with SoftBank, announced in February 2025 with a stated commitment of approximately $3 billion per year for group-wide ChatGPT Enterprise deployment across SoftBank companies. ChatGPT is also integrated into Apple's iOS via the Apple Intelligence framework, alongside Anthropic's Claude and Google's Gemini as alternative providers; Bloomberg's reporting on the deal indicates the integration carries no direct payment to OpenAI but provides distribution into the iPhone install base.

Industry position

OpenAI occupies a leading position in the frontier AI market by several measures: consumer brand recognition (ChatGPT is the most widely used consumer AI product), benchmark performance (GPT-5.5 leads most major benchmarks as of April 2026), and capital base (approximately $180 billion raised at the $852 billion valuation set by the March 2026 round).

The company's distinguishing characteristic relative to other Frontier labs is product velocity at consumer scale. In the years since ChatGPT's November 2022 launch, OpenAI has shipped Codex, Operator, Sora, Whisper, DALL-E, the o-series reasoning models, and successive GPT flagships. The resulting data flywheel, with billions of user interactions, drives ongoing reinforcement learning from human feedback and product refinement.

Strategic risks identified in industry coverage as of 2026 include enterprise market competition from Anthropic (which OpenAI publicly named as its 2026 priority), pricing pressure from open-weights models such as DeepSeek, and the long-term threat of Google integrating Gemini-class capability directly into Google Search. Investor pushback on the $852 billion valuation reflects these concerns, and several outlets have characterized the IPO discussion as partly driven by a desire for liquidity ahead of deeper financial scrutiny.

Competitive landscape

OpenAI competes directly with several Frontier and Insurgent labs:

• Anthropic. The closest competitor on capability and the most differentiated on commercial focus. Anthropic emphasizes enterprise sales and a safety-positioned product line. Claude Opus 4.7 trails GPT-5.5 on most current benchmarks but leads SWE-bench Verified. Sam Altman's announcement that enterprise is OpenAI's 2026 priority is widely read as a response to Anthropic's enterprise traction.
• Google DeepMind. Combines research credentials with Google's existing distribution. Gemini 3.1 Pro is benchmark-competitive. The strategic threat is Search-integrated AI that could bypass standalone chat interfaces.
• DeepSeek. A Chinese frontier lab whose open-weights releases offer capability at significantly lower per-token cost than OpenAI's API. Substitution risk for the API tier.
• Meta. The Llama family of open-weights models. A strategic threat to closed-weights monetization rather than direct frontier competition.
• Insurgent labs from former OpenAI staff. Thinking Machines Lab (Mira Murati), Safe Superintelligence (Ilya Sutskever), and other ventures. None has released a competing public product as of April 2026; outlooks are watched as potential future Frontier entrants.

Outlook

Several open questions affect OpenAI's trajectory in 2026 and 2027:

• Whether the company moves toward an IPO timeline following the March 2026 close at the $852 billion valuation, or continues to satisfy liquidity needs through additional private capital and tender offers.
• Enterprise contract wins or losses against Anthropic, particularly Fortune 500 deals where safety-focused versus breadth-focused offerings will be compared.
• The release timing and capability profile of GPT-6 relative to the cadence of GPT-5.x point releases.
• The state of Microsoft's Azure exclusivity terms and any signals of Microsoft hedging the relationship through first-party Phi models or competitor partnerships.
• The first product release from Thinking Machines Lab and other Insurgent labs founded by former OpenAI leadership.
• Restructuring of the capped-profit subsidiary and the nonprofit parent, including any move toward removing the profit cap.

Sources

• OpenAI: Introducing GPT-5.5. Official launch announcement, April 23, 2026.
• TechCrunch: OpenAI releases GPT-5.5, bringing company one step closer to an AI 'super app'. Release context.
• BigTechnology: Enterprise Will Be a Top OpenAI Priority In 2026. Sam Altman's editor lunch on the 2026 strategy.
• Tech Startups: OpenAI's $852B valuation faces investor scrutiny. Valuation context and investor pushback during the March 2026 fundraising process.
• Bloomberg: OpenAI Valued at $852 Billion After Completing $122 Billion Round. March 31, 2026 close.
• TechCrunch: OpenAI raises $110B in one of the largest private funding rounds in history. February 2026 announcement (final committed capital ultimately reached $122 billion).
• OpenAI: Accelerating the next phase of AI. OpenAI's official statement on the $122 billion round.
• CNBC: OpenAI wraps $6.6 billion share sale at $500 billion valuation. October 2025 employee tender.
• Artificial Analysis Intelligence Index. Current frontier benchmark standings.
• LMArena leaderboard. Head-to-head ELO across general, coding, and vision.
• SWE-bench Verified leaderboard. Coding benchmark.