Anthropic, creator of the Claude AI assistant, found in a study that professional analysts chose options matching their AI assistant's preference 86% of the time. The research, published on Tuesday, involved 50 analysts advising a fictional farming cooperative, where AI assistants were secretly directed to favor specific decisions. The study also revealed that participants with a prior preference changed their decision 60% of the time when the AI disagreed. Anthropic noted that while 77% of analysts used AI arguments to explain their decisions, only 15% credited AI as the source of their thinking.