reddit上有网友在讨论为什么harvey的用户只是尝鲜,但很难长期使用?
有一个高票答案是:
“The problem with AI tools is that you can never really trust the output. Most explicitly say “you must review the result because AI can generate errors. Seniors and partners end up treating it like the work product of a 1st year who never learns anything, which is of course something they love 🙄”
人工智能工具的问题在于,你永远无法真正信任其输出。大多数工具明确表示“你必须审查结果,因为人工智能可能会产生错误。”高级员工和合伙人最终将其视为一个一年级学生的工作成果,而这个学生从未学到任何东西,这当然是他们所喜欢的🙄
对于很多专业场景,不仅追求效率,也要追求准确性,好工具的价值在于提高了内容输出的质量。