OpenAI is a leading American research organization in the field of Artificial intelligence (AI), headquartered in San Francisco, California. Its overarching goal is to develop a safe and useful General Artificial Intelligence (AGI - Artificial General Intelligence), that benefits humanity as a whole. AGI is defined as a highly autonomous system that outperforms humans in most economically valuable activities.
Foundation and corporate structure
OpenAI was originally founded in December 2015 as a non-profit organization (OpenAI, Inc.). Notable founders include Sam Altman, Elon Musk, Ilya Sutskever and Greg Brockman, who started with a $1 billion pledge to advance AI research and development. The company's original mission was to openly share AI research to foster broad collaboration and ensure that AI is not constrained by purely commercial interests.
In 2019, OpenAI made a strategic shift and introduced a for-profit subsidiary, OpenAI Global, LLC. This „capped-profit“ model allows the company to attract large investments required for capital-intensive AI research, talent recruitment and supercomputer development, while maintaining control with the non-profit parent company. The financial returns for investors are capped; all profits in excess of this cap flow back into the non-profit organization for the benefit of humanity. Microsoft is a major investor and has invested over 13 billion US dollars in OpenAI, giving the company access to Azure cloud resources.
Key products and research
OpenAI has contributed significantly to the current AI boom through its research work and product developments. The best-known innovations include:
- ChatGPT: A report published in November 2022 AI chatbot, which is based on the GPT family (Generative Pre-trained Transformer) from large language models (LLMs) based. It is known for its ability to generate human-like text and has significantly spurred public interest in generative AI. Newer versions include GPT-4, GPT-4o and the recently (October 2025) introduced ChatGPT-5.
- DALL-E: These models, which are able to generate detailed digital images from text descriptions, have been under development since 2021.
- Sora: An advanced text-to-video model that generates high-quality videos from textual input.
- OpenAI APIAn interface that gives developers access to various OpenAI models, including GPT-4 and DALL-E, and allows them to integrate these technologies into their own applications.
- OpenAI Gym: A toolkit for research in the field of reinforcement learning published in 2016.
- OpenAI Atlas: A browser with integrated ChatGPT launched in October 2025.
In addition, OpenAI is continuously developing new models such as OpenAI o1, o3 and o4-mini, which offer improved reasoning capabilities and efficiency in different application areas. OpenAI's research covers various aspects of AI, including Deep learning, reinforcement learning and the Natural language processing.
