Jack Clark, co-founder of Anthropic, said at a recent Aspen Institute event that concerns about AI making people less able to think and reason independently are issues AI companies should address.
Clark cited Anthropic’s Claude chatbot as an example: the model has begun asking users clarifying questions about their prompts. He described this behavior as "forcing a person to engage their brain more and claim some agency back." Clark added that future chatbot iterations might even withhold responses until users contribute their own original ideas, but clarified that such a feature is not currently part of Anthropic’s product plans.
He argued that prompting people to work through problems can help stimulate critical thinking. At the same time, Clark noted it is unlikely that users broadly want additional friction introduced into their user experience.
In short, Clark suggested that responsible AI design can include mechanisms that encourage human reasoning rather than supplant it, while balancing the preference for a smooth interaction.
