As a result, OpenAI has developed a technique to align their language models with the instructions given by a user, so they can generate precise, truthful outputs.
