As the field of artificial intelligence continues to develop, how we interact with and understand generative AI models becomes increasingly important. A key aspect is the difference between “briefing” and “prompting” when using generative AI. While the terms may seem interchangeable, there are reasons why we may want to use “briefing” instead of “prompting.”
- From instruction to collaboration: “Prompting” suggests a one-way interaction where the AI is only given instructions, like an order to follow. “Briefing” implies a more collaborative approach between humans and machines. The focus is on providing contextual information and background details. We are changing our perspective: from seeing AI as a passive tool to considering AI as an active collaborator working alongside human input.
- Contextual insight: The challenge and opportunity of LLMs is that a machine can understand the context and generate meaningful responses. “Briefing” means providing context to the AI becomes essential. For example, by providing a comprehensive briefing, ChatGPT can understand the nuances of the current task and generate responses that align with what a human would produce.
- Ethical and moral aspects: Using “briefing,” we are closer to thinking through ethical and moral considerations. This is essential to ensure that AI-generated content aligns with society’s values.
- Intuitive and natural interactions: The term “briefing” suggests a conversational and intuitive interaction, a more natural way to communicate with AI models. “Briefing” opens the door for user-friendly interfaces that are more “human-centered,” thereby creating a broader use of AI technologies.
- AI as a creative partner: If we start to see AI as a creative partner instead of a rigid tool, we can take advantage of its ability to generate ideas and develop thoughts, which leads to more innovative solutions and discoveries. Plus, it also implies a connection to how we have always worked with creative work: briefing has been part of the advertising and communication processes for many years.
- Bias: Bias is a concern. Through a “briefing” approach, a more transparent and reflective process is encouraged. We are aware of the biases that may exist in the data used to train AI models and are aware of how they can affect the results. A thoughtful “brief” can reduce bias and add additional perspectives.
Overall, I think that changing from “prompting” to “briefing is a good idea:
- It can help to clarify the expectations of both the user and AI.
- It can help ensure that AI generates content relevant to the user’s needs.
- It can help to improve the accuracy and quality of AI’s output.
So why not switch from the technical IT-oriented prompting to the more communication-oriented briefing an AI?