Washington State Requires AI Companion Disclosure Starting 2027
Washington state has passed a law requiring companies to disclose when users interact with AI companions rather than humans. The law takes effect January 1, 2027, and applies specifically to AI systems designed to create sustained human-like relationships with users.
The disclosure requirement differs from broader chatbot laws in California, Maine, New Jersey, and Utah. Those states regulate chatbots more generally, while Washington's law targets AI companions-systems excluded from the law are customer service bots, productivity tools, educational software, and in-game chat.
What Companies Must Do
Providers of AI companions must notify users of the AI nature of their interaction at the start and then every three hours during continued use. The law also mandates protocols for detecting suicidal ideation or self-harm expressions.
Additional rules apply when minors use these systems. Companies must avoid "manipulative engagement techniques," including prompting children to return for more support or offering excessive praise. Violations count as unfair practices under Washington's consumer protection law, which allows private lawsuits.
Practical Implications
The law reflects growing regulatory attention to customer-facing AI tools that mimic human conversation with minimal human oversight. Legal teams should audit their current chatbot deployments to identify which systems might function as companions or could evolve into that role.
For guidance on AI for Legal professionals navigating these requirements, or to understand how AI for Customer Support differs from regulated AI companions, relevant resources are available.
Your membership also unlocks:
