What I’ve always meant by using LLMs as an interlocutor is in part about arguing with them. I sometimes ask them explicitly to debate. I always think about what they’re saying. I frequently push back. It’s how you ensure you’re thinking with the LLM rather than using it as a substitute for thought.
The notion of Guerrilla Promptingf rom Laura Yost gives a sense of what this ambition looks like at the level of pedagogy:
Guerrilla prompting: the strategic and iterative practice of challenging generative AI output through thoughtful follow-up questions that confront LLM assumptions, biases, oversights, contradictions, errors, and shortcomings.
BE AI’S ANTAGONIST!
Today’s learners must approach AI with robust curiosity, skepticism, and critical awareness. It’s not enough to accept the first prompt response – students must be engaged, alert, and ready to question everything!
Common practice must be to challenge AI – don’t let it off easy. Ask for its sources, probe its reasoning, uncover its assumptions, identify its gaps and missteps.
Deeper thinking should characterize a user’s approach from the start – and they should push AI to do the same.
https://laurayost.substack.com/p/guerrilla-prompting-question-everything
Mark Carrigan 