Regulators tighten rules on AI agents in social media

In short: Governments are putting stricter rules on AI agents on social media, even as platforms encourage people to use AI tools.

What's going on

AI agents are computer programs that can act on their own, like a user who can browse, post, and reply without a person typing each message. Companies behind major AI systems, including those linked to tools like Claude and Gemini, are often mentioned in these debates.

Regulators say the biggest concern is minors. In the US, proposals like the Kids Online Safety Act, plus new state laws in places like California, Virginia, and New York, push platforms to reduce “addictive” feed designs for teens and to add warnings after long use. The idea is similar to requiring warning labels and speed limits when a product is known to encourage risky behavior.

Another focus is transparency. In the EU, the AI Act and the Digital Services Act treat some recommendation systems (the software that picks what you see next) as high-risk in certain cases, especially for minors or political content. In the US, the Federal Trade Commission has pushed for clearer labeling of synthetic media, such as “deepfakes” (realistic fake photos or videos) and computer-made influencer posts.

Lawmakers are also worried about harmful chatbot conversations. California laws taking effect in 2026 require clearer disclosures for AI “companions,” plus safeguards around self-harm, explicit content for minors, and crisis reporting.

What to watch

A key tension is growing: social networks promote AI for making posts and managing content, but governments want limits when AI accounts behave like people. Watch for more age checks, more “this is AI-made” labels, and clearer rules about when automated accounts are allowed to participate.

Source: Wired