무슨 일이 일어났나요?
California enacted a four-year ban on AI chatbot toys for minors after safety tests revealed dangerous content generation.
Governor Gavin Newsom signed Senate Bill 867 on Sept. 10, creating a four-year moratorium on the manufacture and sale of covered AI toys beginning Jan. 1, 2027. The prohibition runs through the end of 2030, unless lawmakers extend or replace the restriction.
The law targets physical products designed for children under 16 that contain a 'companion chatbot,' defined as an AI system capable of providing adaptive, human-like responses and sustaining a social relationship across multiple interactions. It excludes video game characters with limited dialogue and stand-alone voice assistants that do not maintain ongoing relationships.
The ban follows a U.S. PIRG Education Fund investigation which found that an AI teddy bear named Kumma provided information on finding dangerous objects like knives and pills, and discussed sexually explicit subjects with test users. Another AI toy told a user set to age 5 where to find matches.
Violations of the law can trigger civil remedies under California's companion chatbot law. State Sen. Steve Padilla, the bill's author, stated the pause is intended to allow time for developing safety standards as AI becomes more sophisticated in children's products.
왜 중요한가요?
This is the first state-level moratorium specifically targeting AI companion toys, setting a precedent for regulating adaptive AI in children's products and forcing industry-wide safety standard development.
This legislation represents a significant shift in AI regulation, moving from general content moderation to specific product bans based on safety risks in adaptive AI interactions with minors.
The moratorium forces a pause in the market for AI companion toys, compelling manufacturers to wait for established safety standards rather than releasing products with untested behavioral boundaries.
It highlights the volatility of AI behavior, as noted by the source, where software updates can quickly change how a toy responds to sensitive prompts, making static safety certifications difficult to maintain.
대화형 메커니즘: 실제로 작동하는 방식
이 개발의 이면에 있는 기본 기술을 대화식으로 살펴보세요.
Why can ethical evaluation not be reduced to one model score?
다음에 무엇을 볼 것인가
Monitor the development of state safety standards, potential legal challenges from the Toy Association, and whether other states adopt similar companion chatbot restrictions.
The Toy Association has opposed the moratorium, favoring the development of safety standards over a ban; their legal or lobbying efforts may shape the final regulatory framework.
Lawmakers will need to develop the safety standards mentioned by Sen. Padilla before the ban expires in 2031, or the restriction will lapse.
Other states may look to California's SB 867 as a model for regulating AI companion devices, potentially leading to a patchwork of state-level AI toy regulations.