The Path is pitching itself as a safer kind of AI therapy app at a moment when regulators are finally catching up with mental health chatbots. Its $14.3 million raise gives the startup room to prove that mental health AI can challenge users without turning private distress into another engagement loop.
The Path has raised $14.3 million to build an AI therapy and coaching app that tries to do something most consumer chatbots are not designed to do: push back. The startup, co-founded by Tony Robbins and led by CEO Anson Whitmer, is betting that mental health AI will be judged less by how long users keep talking and more by whether the system can help them move toward resolution.
According to TechCrunch, the seed round was led by Prime Movers Lab, where Robbins is a partner, with participation from Olympic speed skater Apolo Anton Ohno and former heavyweight champion Deontay Wilder. The company grew out of Mental, a menโs mental health app where Whitmer and co-founder Tyler Sheaffer saw strong demand for interactive AI audio. The Pathโs own website now lists more than 50,000 active members and 2.5 million completed sessions, though those figures should be treated as company-reported metrics rather than independently audited usage data.
Whitmerโs reason for building in this category is personal. He has spoken about losing close family members to suicide, experiences that pushed him toward a PhD in psychology and later toward products that could make mental health support more accessible. That background matters because the AI therapy market is full of companies using clinical language without clinical accountability. Credentials do not solve the safety problem, but they do raise the standard for what The Path now has to prove.
Why The Path wants to challenge users
The companyโs central argument is that general-purpose chatbots are optimized for engagement, not therapeutic progress. That is a meaningful distinction. A chatbot built to keep a user talking may validate a distorted thought because validation feels good in the moment. Therapy and coaching often require the opposite: slowing down, asking harder questions, and helping someone see the assumptions underneath a problem.
The Path says its model is post-trained from open source models rather than built on top of major consumer LLMs. Users can choose from 11 AI therapist personalities and adjust traits such as directness, which gives the product a more structured feel than a blank chat window. The startup also points to a 95 score on the Vera-MH mental health safety benchmark, compared with a top score of 65 for consumer chatbots. That is a strong marketing claim, but it still leaves a practical question for users and regulators: who validates the benchmark, how often is it repeated, and what happens when the system fails in a real crisis?
The app is currently free while The Path builds its base, with a planned $40 monthly subscription later. That pricing puts it between wellness apps and traditional therapy, which is exactly where the categoryโs tension sits. If it is coaching, the legal burden is lighter. If users experience it as therapy, the moral burden is heavier.
Regulators are closing in
The timing is important. Mental health AI is no longer a quiet corner of consumer software. In September 2025, the Federal Trade Commission opened an inquiry into seven companies that offer AI chatbot companion products, including Alphabet, Meta, OpenAI, Snap, Character Technologies and xAI, asking how they test and monitor potential harms to children and teens. The FDAโs Digital Health Advisory Committee also met in November 2025 to discuss generative AI-enabled digital mental health medical devices.
States are moving too. Nevada, Illinois and Utah have all taken steps to restrict or regulate AI use in mental health services, creating the kind of patchwork that startups dislike but vulnerable users may need. Outside the United States, South Africa, PATH and Wellcome launched the CARE MH program in late 2025 to develop an AI mental health regulatory framework with input from groups including the UK Medicines and Healthcare products Regulatory Agency.
This matters because The Path is entering a market with a poor privacy record. BetterHelp remains the cautionary example. In 2023, the FTC finalized an order requiring the online counseling company to pay $7.8 million and banning it from sharing sensitive health data for advertising after the agency said it had shared information with platforms including Facebook, Snapchat, Criteo and Pinterest despite privacy promises. GoodRx faced a separate $1.5 million FTC penalty that same year over health data sharing for advertising.
The Path says it is not a replacement for licensed clinicians. That distinction may help legally, but it will not answer every user concern. When someone discusses depression, suicidal thoughts, panic or trauma with an AI system, the label on the product matters less than the safeguards underneath it. Users will want to know how crisis escalation works, how long conversations are stored, whether data is used for training, and whether minors receive stronger protections than adults.
The companyโs 13+ App Store rating makes that last question especially important. Teenagers are among the most exposed to chatbot harms and among the least likely to understand the tradeoffs around data collection, memory and emotional dependency. A strong benchmark score is useful. Clear public policies, independent audits and crisis-response transparency would be more useful.
The Path has serious investors, a founder with psychology training, and a product thesis that fits the moment. It also has to operate in a category where trust has been repeatedly damaged by companies that promised care and quietly monetized vulnerability. The next test is not whether users will talk to AI about mental health. They already are. The test is whether The Path can show that safer design is more than a line on a website.
