This document reflects our strategy that we have refined over the past two years, including feedback from many people, both inside and outside of OpenAI. The timeline for AGI remains uncertain, but our Charter will guide us throughout human development with the best interests of humanity in mind.
OpenAI's mission is to ensure that artificial general intelligence (AGI) – and by that highly autonomous systems that surpass humans in the most economically valuable jobs – benefits all of humanity. We will try to build a safe beneficial AGI directly, but if our work helps others achieve this outcome, we will also consider our mission accomplished. To this end, we are committed to the following principles:
Benefits of being widely distributed
We commit to using any influence we have over the deployment of AGI to ensure that it is used for the benefit of all, and to avoid using AI or AGI to harm humans or overly centralize power.
Our first fiduciary duty is to humanity. We anticipate that significant resources will need to be mobilized to accomplish our mission, but we will always strive to take action to minimize conflicts of interest between employees and stakeholders that could harm a wide range of interests.
Long-term security
We are committed to conducting the research needed to make AGI safe and to drive widespread adoption of such research by the AI community.
We are concerned that the late AGI will develop into a competitive race where there is no time to take adequate safety precautions. Therefore, if a project with the same values and a strong sense of security comes close to building AGI before us, we commit to stop competing with the project and start assisting the project. We'll work on a case-by-case basis, but a typical trigger might be "the odds of success in the next two years are even greater."
Technical leadership
To effectively respond to AGI's impact on society, OpenAI must be at the forefront of AI capabilities – policy and safety advocacy alone are not enough.
We believe that AI will have a broad societal impact before AGI and we will strive to take a leadership role in areas that are directly aligned with our mission and expertise.
We will actively collaborate with other research and policy institutions; We seek to build a global community that addresses AGI's global challenges together.
We are committed to providing public goods and helping society on the path to AGI. Today, this includes publishing most of our AI research, but we anticipate that safety and security concerns will reduce our traditional postings in the future, while increasing the importance of shared security, policy, and standards research.