When most people hear the name OpenAI, they immediately think of ChatGPT. But ChatGPT is not the company itself. It is one product within a broader ecosystem of AI models, developer services and research covering text, images, audio, coding and reasoning.
OpenAI describes itself as an AI research and deployment company. Its stated mission is to ensure that artificial general intelligence benefits all of humanity. That is the organisation’s declared objective, but it should not be treated as proof that every product or business decision necessarily achieves it.
How did OpenAI begin?
OpenAI was announced on December 11, 2015, as a nonprofit AI research organisation. Its founding team included entrepreneurs and researchers such as Sam Altman, Elon Musk, Greg Brockman, Ilya Sutskever, Wojciech Zaremba and John Schulman.
The original idea was to develop artificial intelligence while prioritising a positive outcome for humanity rather than making financial returns the organisation’s main objective.
Building increasingly capable AI systems, however, requires expensive computing infrastructure, specialised chips, energy, researchers and substantial investment. In 2019, OpenAI created a for-profit subsidiary to raise capital and expand its research and deployment work.
How is OpenAI structured today?
OpenAI currently consists of two connected organisations:
- OpenAI Foundation, the nonprofit entity.
- OpenAI Group PBC, the for-profit public benefit corporation.
The nonprofit foundation controls the commercial group through governance and voting rights. OpenAI Group PBC operates the commercial business, raises capital and develops products and services.
This structure is different from that of a conventional company, but it does not remove commercial interests or investor influence. It represents an attempt to combine a public-benefit mission with the financial demands of developing highly expensive AI systems.
What does OpenAI develop?
OpenAI develops models and products that work with text, images, audio, code, search and multi-step tasks.
Its best-known product is ChatGPT, an AI assistant used for writing, learning, research, analysis, programming and content creation. The company also operates the OpenAI API, which allows developers and organisations to integrate OpenAI models into their own applications and services.
It is important to distinguish between three different things:
- OpenAI is the company.
- ChatGPT is a user-facing product.
- GPT and other model families are the technical systems that may power ChatGPT or be accessed through the API.
A single product can use several models, while the same model may be available through more than one product.
What does artificial general intelligence mean?
OpenAI uses the term artificial general intelligence, or AGI, to describe systems that go beyond narrow, specialised tools and can perform a wide range of intellectual tasks at a high level.
There is no universally accepted definition of AGI and no single test that can conclusively establish that a system has reached it. Claims about achieving AGI should therefore be examined carefully, with a clear distinction between measurable improvements in AI capabilities and promotional language or predictions about the future.
How does OpenAI make money?
OpenAI generates revenue through paid ChatGPT subscriptions for individuals, businesses and institutions, as well as usage of the OpenAI API. It also receives capital through commercial partnerships and investment.
Product prices, model availability and usage limits can change frequently. Anyone making a purchasing or integration decision should consult OpenAI’s current pricing pages instead of relying on figures quoted in a static article.
Is OpenAI open source?
The company’s name can create confusion. OpenAI has published research, software and models with different levels of openness throughout its history, but its main products and frontier models are not all open source.
Making a model available through ChatGPT or an API does not mean that its source code, model weights or training data are publicly available. Each model or tool must be evaluated according to its specific licence and documentation.
What are the risks surrounding OpenAI’s products?
OpenAI’s models can produce incorrect information, reflect biases present in data or be misused to create misleading content. These technologies also raise questions about privacy, copyright, employment, automation and the concentration of technological power among a small number of companies.
OpenAI says it uses testing, evaluations, safeguards and human oversight to reduce risks. The existence of safety measures, however, does not guarantee that errors or misuse will disappear.
Users and organisations therefore retain responsibility for how these systems are used. Sensitive information should not be submitted without understanding the relevant data policy, important outputs should be verified, and general-purpose AI should not replace qualified professional judgement in consequential medical, legal or financial decisions.
What is the difference between OpenAI and ChatGPT?
OpenAI is the organisation that conducts research and develops AI models and products. ChatGPT is an assistant and interface built using some of the company’s technology.
A company is not the same thing as its product, and a product is not necessarily the same thing as the model powering it. OpenAI is the company, ChatGPT is the product, and the underlying model can change as the service evolves.
Conclusion
OpenAI is an AI research and deployment company that began as a nonprofit organisation in 2015. It now operates through the nonprofit OpenAI Foundation and the commercial OpenAI Group PBC.
The company develops ChatGPT, the OpenAI API and several families of AI models. Its products have helped make generative AI accessible to a broad audience, but OpenAI remains a powerful commercial and technical organisation whose claims and products should be assessed through evidence and measurable outcomes, not its stated mission alone.