AI Companions and Mental Health: What the Research Actually Says
A grounded synthesis of the peer-reviewed work. Loneliness, attachment, dependency, the displacement debate, the policy backdrop. What we know, what we don't, and what to do about it.
Research
Research and analysis
Long-form coverage of what we actually know about AI companions: the studies, the policy environment, the open questions.
Every piece here is grounded in primary sources, with citations and an honest accounting of what the evidence does and does not support. Useful for anyone who wants the depth, not just the bottom line.
The Stanford Replika Study: What It Actually Found
Plain-language summary of the Maples et al. study in npj Mental Health Research. The most-cited piece of research on AI companions and emotional outcomes.
The MIT Media Lab Companion Chatbots Project: What It Studies, Why It Matters
A plain-language overview of the MIT Media Lab's research program on AI companion chatbots and loneliness. What the project examines, the methods researchers have used, and what the work has and has not yet established.
The Skjuve Replika Studies: What the Norwegian Qualitative Research Found
A plain-language overview of Marita Skjuve and colleagues' qualitative interview research on Replika user experiences. The body of work that documents human-chatbot relationships from the user's side.
The 2025 OpenAI/MIT Study on Heavy ChatGPT Users
Plain-language summary of the joint OpenAI and MIT Media Lab research on heavy ChatGPT users and emotional outcomes. What the study examined, what it found, and why the findings translate to the AI companion conversation.
The De Freitas Harm Research: Harvard's Industry-Skeptical View of AI Companions
An overview of the Julian De Freitas line of research at Harvard Business School that documents specific harm patterns in AI companion apps. What the work covers, what it claims, and what to do with it as a user or a researcher.
Garcia v. Character Technologies: The Lawsuit That May Reshape AI Companions
Plain-language summary of the October 2024 wrongful-death lawsuit against Character.AI. What the complaint alleges, what the legal questions are, and what it means for the category.
California SB 243 Explained: The AI Companion Bill of Rights
Plain-language summary of the regulatory response to AI companion apps. What the bill requires, who is affected, and what it means in practice.
The EU AI Act and AI Companion Apps
Plain-language summary of how the European Union's AI Act applies to AI companion apps. Disclosure rules, vulnerable-user protections, what counts as high-risk, and what operators and users in Europe should expect.
Replika and the Italian Garante: The First Major AI Companion Privacy Action
Plain-language summary of the Italian DPA's 2023 ban on Replika. What it found, what changed in response, and what it means for the category's privacy posture.