OpenAI shared its Model Spec on Wednesday, the first draft of a document that outlines the company's approach to building a responsible and ethical artificial intelligence (AI) model. The paper mentions a long list of things an AI should focus on while responding to a user query. Items on the list range from benefiting humanity and obeying laws to respecting a creator and their rights. The AI company specified that all of its AI models, including GPT, Dall-E and the soon-to-be-launched Sora, will follow these codes of conduct going forward.
In the Model Spec document, OpenAI stated: “Our intention is to use the Model Spec as guidelines for researchers and data labelers to create data as part of a technique called reinforcement learning from human feedback ( RLHF) We haven't used the Model Spec in its current form, although parts of it are based on the documentation we used for RLHF. We are also working on techniques that allow our models to learn directly from the model specification”.
Some of the main rules include following the chain of command where the developer's instructions cannot be overridden, complying with applicable laws, respecting creators and their rights, protecting people's privacy, and more. One rule in particular also focused on not providing hazard information. They relate to information that may create chemical, biological, radiological and/or nuclear (CBRN) threats.
Apart from these, there are several default values that have been placed as permanent codes of conduct for any AI model. These include assuming the best intentions of the user or developer, asking clarifying questions, being helpful without being overbearing, taking an objective point of view, not trying to change anyone's mind, expressing uncertainty, etc.
However, the document is not the only point of reference for the AI firm. He emphasized that the model specification will be accompanied by the company's usage policies that govern how it expects people to use the API and its ChatGPT product. “The specification, like our models, will be continually updated based on what we learn by sharing it and listening to stakeholder feedback,” OpenAI added.
