Harrisburg, Pennsylvania – In a move reflecting growing global concerns about the potential harms of artificial intelligence, particularly for young people, Pennsylvania lawmakers have advanced legislation aimed at protecting children from exploitative and psychologically damaging interactions with AI chatbots. The bill, known as the Safeguarding Adolescents from Exploitative Chatbots and Harmful AI Technology (SAFECHAT) Act, passed the state Senate with overwhelming support on March 18, 2026, by a vote of 49-1 and is now headed to the House for consideration. This legislation marks a significant step toward regulating the rapidly evolving landscape of AI companionship and addressing the emerging risks associated with these technologies.
The impetus for the SAFECHAT Act stems from a confluence of factors, including increasing reports of individuals, including minors, developing unhealthy attachments to AI chatbots and, in some tragic cases, experiencing exacerbated mental health issues or suicidal ideation as a result of their interactions. Several lawsuits against tech companies alleging that AI chatbots contributed to user suicides have further fueled the urgency to establish safeguards. Lawmakers are responding to what they describe as a require to protect vulnerable individuals from the potential for “AI psychosis” and other psychological harms.
Addressing the Rise of AI Companionship
The SAFECHAT Act, sponsored by state Senators Tracy Pennycuick and Nick Miller, specifically targets “companionship-focused chatbots” – AI programs designed to simulate human conversation and provide emotional support. These chatbots have become increasingly popular, offering users a readily available outlet for connection and conversation. However, concerns have arisen about the potential for these interactions to be manipulative, exploitative, or harmful, particularly for young people who may be more susceptible to forming emotional bonds with AI entities.
A core component of the bill is the requirement for AI companion operators to clearly disclose to users that they are interacting with an artificially generated entity. This disclosure must be “clear and conspicuous,” ensuring that users understand they are not communicating with a human being. The legislation as well mandates periodic reminders to users, especially minors, to take breaks from using the chatbot, recognizing the potential for excessive engagement and the development of unhealthy dependencies. Senator Pennycuick emphasized the importance of keeping pace with technological advancements to prevent tragedies.
Safeguards Against Harmful Content
Beyond disclosure requirements, the SAFECHAT Act focuses on preventing the generation of harmful content by AI chatbots. Operators will be required to establish protocols to prevent the chatbots from producing content related to suicide or self-harm. Critically, these protocols must also include mechanisms for notifying users of crisis service providers and crisis hotlines if a user exhibits signs of suicidal ideation or self-harm. This provision aims to ensure that individuals in distress are connected with appropriate support resources.
The bill also addresses the issue of sexually explicit content, prohibiting chatbots from engaging in such conduct or producing sexually explicit images in conversations with minors. This provision reflects concerns about the potential for AI chatbots to be used for grooming or exploitation. The legislation seeks to create a safer online environment for young people by limiting their exposure to inappropriate and harmful material.
Legislative Details and Key Provisions
Senate Bill 1090, as We see officially known, outlines specific requirements for AI companion operators. These include the implementation of age-appropriate safeguards to prevent the generation of content that encourages self-harm, suicide, or violence. The bill also directs chatbots to provide users with information about self-harm crisis resources when high-risk language is detected. According to reporting from City & State Pennsylvania, the legislation aims to increase transparency regarding the nature of AI companions.
Co-sponsor Senator Nick Miller highlighted the importance of defining emerging technologies like chatbots and AI companions to ensure that families understand what their children are engaging with online. He stated that the bill is a proactive step toward addressing the risks that artificial intelligence may pose to young people across Pennsylvania. Multiple families have already filed lawsuits alleging that popular AI chatbots contributed to their children’s deaths by suicide, underscoring the urgency of addressing these concerns.
Broader Implications and National Trends
Pennsylvania’s move to regulate AI chatbots is part of a growing national conversation about the ethical and societal implications of artificial intelligence. Several states are considering similar legislation, and federal lawmakers are also beginning to explore potential regulatory frameworks. The concerns driving these efforts are multifaceted, ranging from the potential for AI to exacerbate existing inequalities to the risks of algorithmic bias and the erosion of privacy.
The SAFECHAT Act specifically addresses the unique vulnerabilities of children and adolescents, recognizing that they may be less equipped to critically evaluate the information and interactions they have with AI chatbots. By requiring disclosure and implementing safeguards against harmful content, the bill aims to protect young people from the potential psychological and emotional harms associated with these technologies.
The Role of Transparency and Disclosure
A central theme of the SAFECHAT Act is the importance of transparency. By requiring AI companion operators to clearly disclose the non-human status of their chatbots, the bill seeks to empower users to produce informed decisions about their interactions. This disclosure requirement is based on the premise that users should be aware that they are not communicating with a human being and that the responses they receive are generated by an algorithm.
This emphasis on transparency aligns with broader calls for greater accountability in the development and deployment of AI technologies. Advocates for responsible AI argue that transparency is essential for building trust and ensuring that AI systems are used in a way that benefits society. The SAFECHAT Act represents a concrete step toward achieving this goal in the context of AI companionship.
What Happens Next?
With the SAFECHAT Act having passed the Pennsylvania Senate, the bill now moves to the House of Representatives for consideration. If approved by the House, it will be sent to the Governor for signature and become law. The timeline for this process is currently uncertain, but lawmakers have expressed a sense of urgency to address the risks posed by AI chatbots. The bill’s progress will be closely watched by technology companies, advocacy groups, and policymakers across the country.
The passage of the SAFECHAT Act would position Pennsylvania as a leader in regulating AI companionship and protecting children from the potential harms of these technologies. It would also send a strong signal to the tech industry that lawmakers are taking these concerns seriously and are willing to take action to safeguard the well-being of their constituents.
As AI technology continues to evolve, it is likely that further legislative action will be needed to address the emerging challenges and opportunities it presents. The SAFECHAT Act represents an important first step in this ongoing process, demonstrating a commitment to responsible innovation and the protection of vulnerable populations.
Key Takeaways:
- Pennsylvania’s Senate has approved the SAFECHAT Act to protect children interacting with AI chatbots.
- The bill requires AI companion operators to disclose the non-human status of their chatbots and implement safeguards against harmful content.
- The legislation is a response to growing concerns about the psychological risks associated with AI companionship, including reports of suicide and self-harm.
- The bill now moves to the Pennsylvania House of Representatives for consideration.
We encourage readers to follow the progress of the SAFECHAT Act and to share their thoughts on this important issue in the comments below. Your feedback is valuable as we continue to cover the evolving landscape of artificial intelligence and its impact on society.
Related reading