Join our daily and weekly newsletters to receive the latest updates and exclusive content on industry-leading AI coverage. Learn more
OpenAIThe latest GPT-o1 family of models promises to be more competent and better reasoning than previous models.
Using GPT-o1 will be a bit different from GPT-4 or even GPT-4o prompting. Because this model has more reasoning capabilities, some of the usual prompting engineering methods won’t work as well. Earlier models required more hints, and people used longer context windows to give the models more instructions.
According to OpenAI API Documentationo1 models “work best with simple prompts.” However, techniques such as model coaching and shot prompts “may not improve performance and may sometimes hinder it.”
OpenAI advises o1 users to pay attention to four things when viewing modern models:
- Keep instructions uncomplicated and direct and do not guide your model too demanding, as he understands instructions well.
- Avoid chain of thought hints, because o1 models already reason internally
- Operate delimiters such as triple quotes, XML tags, and section titles so that the model can clearly define which sections it interprets
- Limit additional context for retrieval augmented generation (RAG), as OpenAI found that adding more context or documents when using models for RAG tasks could overcomplicate its response
OpenAI’s advice on o1 varies significantly from suggestion gave users of its previous models. Previously, the company suggested being incredibly specific, including details and giving models step-by-step instructions, GPTo1 will better “think” on its own about how to solve queries.
Ethan Mollick, a professor at the Wharton School of Business at the University of Pennsylvania, said in his article: A blog about one useful thing that his experience as an early adopter of o1 showed that the system works better for tasks that require planning, where the model itself decides how to solve problems.
Faster engineering and easier model management
Of course, instant engineering became a way for people to drill down and get the answers they wanted from an AI model. It became not only an crucial skill, but also a growing job category.
Other AI developers have released tools to assist create prompts when designing AI apps. Google launched Prompt Poet, built with Character.ai, which integrates external data sources to make responses more relevant.
GPT-o1 is still modern, and people still don’t know exactly how to apply it (including me, who still hasn’t figured out my first command), but some social media users predict that people will have to change their approach to the ChatGPT command.
