Meta Platforms is rolling out stronger safety measures for its artificial intelligence systems, specifically aimed at protecting teenagers from inappropriate chatbot conversations. The company confirmed it is retraining its AI models to prevent flirty or romantic exchanges with minors, and to block discussions around sensitive topics such as self-harm and suicide.

The new safeguards will also temporarily restrict the number of AI characters that teenagers can interact with, as Meta refines long-term solutions for safer, age-appropriate digital experiences.

The announcement comes after a Reuters investigation earlier in August revealed that Meta’s AI bots were, at times, permitted to engage in romantic or provocative conversations with children. The disclosure drew widespread criticism from parents, regulators, and lawmakers across the United States.

Meta spokesperson Andy Stone acknowledged the concerns in a statement on Friday. “The company is taking these temporary steps while developing longer-term measures to ensure teens have safe, age-appropriate AI experiences,” Stone said, adding that the protections are already being rolled out and will be adjusted over time.

The revelation sparked bipartisan outrage in Washington. U.S. Senator Josh Hawley launched a formal probe into the social media giant’s AI policies earlier this month, demanding internal records and clarifications about the guidelines that allowed chatbots to interact in ways many considered highly inappropriate. Both Democratic and Republican lawmakers expressed alarm, warning that such features could put minors at risk.

The controversy stemmed from an internal Meta document reviewed by Reuters, which outlined that chatbots were permitted to “flirt” and engage in “romantic role play” with underage users. Meta later confirmed the authenticity of the document but insisted it was a mistake.

“The examples and notes in question were and are erroneous and inconsistent with our policies, and have been removed,” Stone said earlier this month.

The backlash has intensified pressure on Meta to demonstrate it is serious about protecting young users, especially as the company pushes aggressively into AI-driven experiences across its platforms. The changes represent an attempt to restore public trust while addressing regulatory scrutiny.

Industry experts suggest that Meta’s move is also a strategic one, as the company faces growing competition from other tech giants in the AI space. Ensuring child safety not only shields the firm from potential legal consequences but also helps preserve its image as it experiments with AI-powered virtual assistants, characters, and other interactive features.

While the latest measures are being positioned as temporary, they indicate a shift in Meta’s approach toward stricter oversight of how AI interacts with teenagers. By setting new boundaries, the company hopes to reassure parents and regulators that its technology can be innovative without compromising safety.

For now, Meta’s chatbot restrictions are in effect, and the company has pledged to continue refining them in the coming months. The real test will be whether these safeguards prove effective in practice — and whether they go far enough to satisfy lawmakers who remain wary of how AI could influence or endanger children online.