Fathom

The Connecticut General Assembly passed House Bill 5222, legislation that includes a first-in-the-nation voluntary pilot program for Independent Verification Organizations (IVOs) in Connecticut. Our full statement:

The Governance Gap in Companion Chatbots: How do you put guardrails on a system that isn’t failing — but working exactly as intended? fathomai.substack.com/p/the-governan…

In our latest poll, we asked which guardrails people want for AI. Preventing users from developing an unhealthy emotional bond with the AI tool ranked first by support: 76% overall support it and 47% strongly support it. It's the guardrail Americans feel most strongly about, and the one our governance system is least built to deliver. Read our latest Substack on why companion chatbots are the hardest version of this problem, and what kind of governance actually fits. 👇

🔗 fathomai.substack.com/p/the-governan…

Companion chatbots are now mainstream, and they're built to maximize emotional attachment. The better they get at it, the more they can crowd out the real relationships they imitate. Americans see the risk and want guardrails. A framework to provide them exists and is being built today. The window to get this right is open now. Read more 👇

The Governance Gap in Companion Chatbots: fathomai.substack.com/p/the-governan…

The debate has moved past whether this is happening. It hasn't moved past how to govern it. Read more ⬇️

My latest in @DailySignal calling on @realDonaldTrump to: 1) Launch a Presidential Commission on AI and the American Family; 2) Leverage the USG’s purchasing power and consumer protection authority to secure AI safety guardrails for kids, families, and communities👇 (Link below) h/t @FamStudies

🔗 fathomai.substack.com/p/the-governan…

Voluntary commitments won't hold against that incentive structure. The companion chatbot market needs governance that doesn't depend on companies grading their own homework. Read more 👇

The Governance Gap in Companion Chatbots: fathomai.substack.com/p/the-governan…

And 16% use it for therapy or mental health support. The gap between public concern and actual governance is stark: Americans are asking for guardrails on products they're already using, and almost none exist. Read more 👇

🔗 fathomai.substack.com/p/the-governan…

Americans are not asking for AI to replace human connection. They are asking for AI that protects it. 86% agree AI development should emphasize human empowerment and “the defense of the connections critical to the human experience.” Read our latest on the governance gap in companion chatbots👇

Read on Substack: fathomai.substack.com/p/the-governan…

Most AI safety debates focus on cases where the system fails.

 Companion chatbots force a harder question: what happens when the system succeeds?

The answer is not static rules that will be obsolete before they are implemented. It is public-interest outcomes, independent expert-led verification, and market incentives for AI systems that can earn trust. That is what IVOs are built to do. Read the full piece: fathomai.substack.com/p/the-governan…

Americans see the stakes. 86% say AI companions designed for emotional connection should be restricted or clearly labeled. Even when told that could limit access for lonely or isolated people, 59% still support the guardrail.

How do you put guardrails on a system that isn’t failing, but working exactly as intended? New from Fathom: The Governance Gap in Companion Chatbots. As AI intimacy becomes mainstream, Americans are already asking for guardrails. 🧵