OpenAI

OpenAI started in 2015 as a nonprofit research lab with a mission that sounded almost quaint at the time: make sure artificial general intelligence benefits all of humanity. Then ChatGPT happened, became the fastest-growing consumer app in history, and turned a research outfit into the company that defines an industry.

The structure it grew into is unusual on purpose. A nonprofit parent, the OpenAI Foundation, governs a for-profit public benefit corporation, the OpenAI Group, so the mission has legal teeth over the money. And there is now a great deal of money: a $122 billion round in early 2026 valued the company at $852 billion, and it has confidentially filed a draft S-1, signaling a public offering ahead.

OpenAI builds the frontier AI models and then ships them in forms ordinary people and businesses can pick up and use. ChatGPT is the front door for hundreds of millions of consumers; the API platform is the back door for developers embedding GPT into their own products; and a growing agent layer, built on the Agents SDK, Responses API, and Realtime API, is the bet on what comes after the chat box.

The throughline is deployment, not just research. The company publishes cutting-edge work, but its center of gravity is turning frontier capability into products that resolve support tickets, write and refactor code, generate images and voice, and increasingly take actions across a business's systems.

Underneath the consumer sheen, OpenAI sells a fix for a handful of expensive, universal bottlenecks. Software teams are slow and code is costly to ship. Support queues don't scale. Knowledge work stalls on turning messy data into decisions. And every company that wants AI leverage runs into the same wall: they can't hand sensitive data to a model or give up governance to get it.

The products map cleanly onto those pains, coding and refactoring, autonomous support, research and synthesis, content at volume, and an enterprise control plane that makes all of it safe to deploy at work.

OpenAI sells to three audiences at once: consumers who want a capable everyday assistant, developers who build AI features on the API, and enterprises standardizing on ChatGPT across whole workforces. The enterprise motion is now the fastest-growing part of the business, crossing a million business customers, with real deployments spanning finance, sales, operations, IT, and marketing teams.

The lineup spans consumer, developer, and enterprise. ChatGPT is the flagship assistant across web and mobile; the API Platform exposes the GPT-5.5 family and tooling for developers; Codex is the coding agent that reads and writes files, runs shell commands, and handles GitHub PRs; and the platform's voice, image, and realtime capabilities power everything from support bots to creative tools. The common engine underneath is OpenAI's frontier model family.

OpenAI makes money two ways. Consumers and teams climb a subscription ladder, from a free tier through Go, Plus, Pro, and Business, up to custom Enterprise contracts. Developers pay per token on the API, metered by model and context length.

The free tier is the top of the funnel, not the product: it is deliberately limited to push heavy users toward paid plans, while the API's pay-as-you-go pricing scales with how much intelligence a customer actually consumes.

OpenAI still sets the pace, but the gap it once enjoyed has narrowed. Anthropic's Claude, Google's Gemini, Meta's Llama, Mistral, and xAI all now ship models that trade blows on real work, and enterprise buyers increasingly run more than one.

What keeps OpenAI ahead is less any single benchmark than its position: ChatGPT is the default consumer AI and the distribution lead that comes with it, a mission-locked governance structure rivals don't have, and the capital and compute, an $852B valuation and roughly $180B raised, to fund frontier-scale training few can match. Its bet now is agents and enterprise depth, not just a bigger model.

The adoption is not hypothetical. OpenAI has named enterprise customers including Morgan Stanley, Cisco, Target, Lowe's, T-Mobile, Amgen, Booking.com, and Commonwealth Bank, plus early ChatGPT Enterprise adopters like Canva, PwC, Zapier, Estée Lauder, and Block. Samsung, Uber, and a wave of Korean firms have since signed on. The through-line: these are standardization decisions, not pilots.

OpenAI's reputation in 2026 is more contested than it was at ChatGPT's peak. Fans still praise the polish, the mobile apps, the generous free tier, and a memory-plus-custom-instructions setup that few rivals match for personalization, all wrapped in the deepest ecosystem of custom GPTs and integrations.

The criticism is just as real. Power users complain about quality regressions and more frequent refusals, its assistant market share has slipped from its early dominance, and the move from nonprofit ideals toward aggressive commercialization and defense partnerships has cost it goodwill with part of its original base.

OpenAI has raised roughly $180 billion across its history, and its early-2026 round alone brought in $122 billion at an $852 billion valuation, one of the largest private financings ever assembled. Microsoft remains a strategic backer, but the latest round widened the tent dramatically, with Amazon, NVIDIA, and SoftBank anchoring it alongside a16z, Sequoia, Thrive, T. Rowe Price, BlackRock, Fidelity, and a long list of global institutions.

The confidential draft S-1 on file with the SEC points to where this is heading: a public company, with the financial transparency and governance that entails.

OpenAI's next chapter is about extending a lead under real pressure. The product bet is agentic: systems that don't just answer but act, from Codex in the developer world to Operator-style browser agents. The corporate bet is the public markets, with the draft S-1 signaling an IPO that would force new transparency on a company used to operating on its own terms. Both play out while rivals close the model gap and part of the user base questions the mission drift.

OpenAI hires curious people from a wide range of disciplines and runs on four stated values, Humanity first, Act with humility, Feel the AGI, and Ship joy, backed by operating principles that prize agency (Find a way), creative problem-solving over control, fast updating on new information, and intense focus. Ideas are supposed to come from anywhere regardless of title.

The benefits back the intensity: competitive health and mental-health coverage, 24-week paid birth-parent leave (20 weeks for other parents) with a work-from-home ramp, fertility and family-planning support, a learning stipend, daily meals, and coaching sessions. A six-month Residency offers a path in for engineers and researchers moving into AI. It is a San Francisco-headquartered, in-person-leaning shop hiring across engineering, research, product, and go-to-market.

OpenAI discloses salary bands on a large share of its US roles, and the numbers are frontier-lab high. Engineering base spans roughly $86K to $585K depending on level, research tracks close behind, and even non-technical functions clear $150K at the low end. The twist is equity: instead of conventional stock options, OpenAI grants Profit Participation Units, a share of future profits that has become a signature and much-debated part of its offer. Sales roles carry on-target commission on top of base.

For business customers, OpenAI leans hard on trust: no training on customer data by default, zero-data-retention by request, HIPAA Business Associate Agreements, SOC 2 Type 2, AES-256 at rest and TLS 1.2+ in transit, SSO/MFA, IP allowlisting and mTLS, and data-residency controls across multiple regions. The consumer terms are entered into with OpenAI OpCo, LLC, a Delaware company, and the privacy policy was last updated in May 2026.

Two storylines dominate OpenAI's recent coverage. The first is money, its $122 billion round at an $852 billion valuation drew wall-to-wall attention as one of the largest private financings ever, with the confidential S-1 keeping IPO speculation alive. The second is enterprise traction, from crossing a million business customers to landing marquee names like Samsung, the press is tracking how fast ChatGPT is becoming default infrastructure inside big companies.