Sign up to save tools and stay up to date with the latest in AI
bg
bg
1

AI chatbots for historical figures are an ethical minefield

Oct 01, 2023 - businessinsider.com
Tech companies and AI startups are creating chatbots that mimic historical figures, with some intended to be educational tools for classrooms. However, educators have raised concerns about the accuracy of these bots and their potential to negatively impact critical-thinking skills. The bots, which are trained on datasets from sources like Reddit and Wikipedia, have been found to make errors and sometimes "make up facts," according to Sal Khan, founder of Khan Academy.

Experts suggest that to use these bots ethically, they need defined inputs and a retrieval-augmented approach to ensure historically accurate conversations. Overreliance on these bots could lead to a decline in critical-reading skills and affect students' abilities to assimilate, synthesize, and create new ideas. Some educators believe that while these bots could be used as a starting point, they should not replace active engagement with original source materials.

Key takeaways:

  • Tech companies and AI startups are creating chatbots that mimic historical figures, with some intended for educational use in classrooms.
  • However, these bots often provide inaccurate or misleading information, and their use could lead to a decline in critical-thinking skills among students.
  • AI experts suggest that to use these bots ethically, they need defined inputs and a retrieval-augmented approach to ensure historically accurate information.
  • Despite the potential drawbacks, some educators believe that the bots could help students understand the subjective nature of history and avoid overused ideas in their own work.
View Full Article

Comments (0)

Be the first to comment!