On Monday, Americans for Responsible Innovation (ARI) released a new report exploring chatbot harms to minors and calling on federal agencies to develop new national evaluations examining AI chatbot safety for minors. The report finds that current and proposed AI chatbot safeguards at the federal and state level focus on mitigating minors’ exposure to AI harms via age-gating and parental controls, but that few design-level safeguards exist to ensure the safety of the underlying models.
Read the full report:
Growing Up With Chatbots: Why We Need National Minor Safety Benchmarks
By Brandie Nonnecke, PHD, Allie Maloney, and Megi Llubani
“As a generation of young people increasingly turns to AI chatbots, these new tools pose serious risks related to unhealthy dependency, diminished human connection, and harmful recommendations,” said ARI Senior Policy Director Brandie Nonnecke. “Implementing a national set of minor safety benchmarks would create a race to the top among industry. Our national strategy on AI must encourage and improve model safety before deployment, not just address harms after they occur.”
The new report recommends that the National Institute of Standards and Technology (NIST) and the Center for AI Standards and Innovation (CAISI) develop national minor safety benchmarks that define the harm landscape and establish recommended best practices for embedding minor safety during model development and pre-deployment testing. The authors recommend that CAISI use the national minor safety benchmarks and best practice guidance to evaluate the most widely used AI chatbots by minors in the U.S.
This week’s release follows another legislative proposal from ARI last month, recommending that policymakers require a non-personified default setting for chatbots used by minors.
###
Americans for Responsible Innovation (ARI) is a nonprofit organization dedicated to policy advocacy in the public interest, focused on emerging technologies like artificial intelligence (AI). Learn more at ARI.us.
