Google add mental health tools to Gemini chatbot after lawsuit

WASHINGTON – Alphabet’s Google plans to introduce new mental health support features for its Gemini chatbot as the company and rivals, like OpenAI, have faced several lawsuits accusing their artificial intelligence tools of leading to harm. 

Gemini will add an interface directing chatbot users to a support hotline when the conversation indicates “a potential crisis related to suicide or self-harm”, Google said in a blog post on April 7.

Additionally, the company is adding a “help is available” module for chats about mental health and design tweaks to discourage self-harm. 

The rapid explosion of tools like Gemini and ChatGPT have led to some users forming obsessive relationships with AI bots, allegedly contributing to delusions and, in extreme cases, murder-suicides. Several families have sued leading AI developers over the issue. Congress has looked into potential threats chatbots pose to children and teenagers.  

In March, the family of a deceased 36-year-old man in Florida sued Google, claiming that his use of Gemini culminated in a “four-day descent into violent missions and coached suicide”. At the time, Google said the chatbot referred the man to a crisis hotline many times but promised to improve the tool’s safeguards.

In other instances, chatbot users have said that AI tools convinced them to act on clear falsehoods. In the blog post, Google said it has trained Gemini “not to agree with or reinforce false beliefs, and instead gently distinguish subjective experience from objective fact”. The company did not provide further details on this process.

In the past, Google has made similar adjustments to its popular services after facing scrutiny, adding information from health institutions and professionals to its search engine and YouTube.

Google also said it was donating US$30 million (S$38.5 million) to global crisis support services over the next three years. BLOOMBERG