The basis of philosophy is the ability to communicate ideas clearly and precisely.
According to Anthropic's resident philosopher, this is also the key to getting the most out of your AI models.
Amanda Askell, a member of Anthropic's technical team and a trained philosopher, says creating effective prompts requires striking the right balance between several considerations.
Askel, who studied philosophy at Oxford University and New York University, explained his thought process on Anthropic's “Ask Me Anything” podcast, according to LinkedIn.
“It's very difficult to extract what's going on because one thing is like the willingness to interact with the model a lot and actually observe the output one after the other,” she said.
A good teleprompter must be “very experimental,” she says.
But Prompt goes beyond experimentation, and here her philosophical training came in handy.
“This is actually where I think philosophy can actually help provide a prompt in a way, because a lot of my job is just trying to explain as clearly as possible the issues and concerns and ideas that I want to model,” she said.
Focusing on clarity not only helps people improve their own prompts, but is also important in understanding the AI itself.
In Anthropic's Prompt Engineering Overview published in July, the company said users who interact with Claude, the company's chatbot, should think of it as a “smart but very new employee (with amnesia) who needs clear instructions.”
“Claude has no context for your norms, style, guidelines, or preferred way of working. The more precisely you explain what you want, the better Claude will respond,” Antropic wrote.
Veteran venture capitalist Marc Andreessen said last month that the power of AI comes from treating it as a “thinking partner.”
“Part of the art of AI is what questions you ask it,” he said.
Those who master this skill can immediately land high-paying jobs as engineers, with an average salary of $150,000, according to levels.fyi, a platform for researching and comparing salaries for engineers.
