A cornerstone of philosophy is the ability to communicate ideas clearly and precisely.<\/p>\n
That’s also the key to getting the most out of an AI model, according to Anthropic’s own resident philosopher.<\/p>\n
Amanda Askell, a member of Anthropic’s technical team and a trained philosopher, says that effective prompting<\/a> requires striking the right balance between several considerations.<\/p>\n On Anthropic’s “Ask Me Anything” podcast, Askell, who studied philosophy at Oxford and New York University, according to her LinkedIn, explained her thought process.<\/p>\n “It is really hard to distill what is going on because one thing is just like a willingness to interact with the models a lot and to really look at output after output,” she said.<\/p>\n Good prompters should be “very experimental,” she said.<\/p>\n Prompting goes beyond experimentation, however, and this is where her philosophical training has helped.<\/p>\n “This is where I actually do think philosophy can actually be useful for prompting in a way because, like, a lot of my job is just being like I try and explain some issue or concern or thought that I’m having to the model as clearly as possible,” she said.<\/p>\n That emphasis on clarity is important not just to help people refine their own prompts but also in understanding AI itself. <\/p>\n In a “Prompt Engineering Overview” that Anthropic published in July, the company said users interacting with Claude, its chatbot, should think of it as “a brilliant, but very new employee (with amnesia) who needs explicit instructions.”<\/p>\n “Claude does not have context on your norms, styles, guidelines, or preferred ways of working. The more precisely you explain what you want, the better Claude’s response will be,” Anthropic wrote.<\/p>\n