Welcome to the forefront of conversational AI as we explore the fascinating world of AI chatbots in our dedicated blog series. Discover the latest advancements, applications, and strategies that propel the evolution of chatbot technology. From enhancing customer interactions to streamlining business processes, these articles delve into the innovative ways artificial intelligence is shaping the landscape of automated conversational agents. Whether you’re a business owner, developer, or simply intrigued by the future of interactive technology, join us on this journey to unravel the transformative power and endless possibilities of AI chatbots.
Daily e-Edition
Evening e-Edition
Sign up for email newsletters
Sign up for email newsletters
Daily e-Edition
Evening e-Edition
Trending:
Shizhen Jasper Jia watched his mother and grandmother care daily for his grandfather as he battled Alzheimer’s disease.
Frequently his grandfather was in a bad mood and he consistently needed round the clock care, Jia said.
Jia could see the emotional toll wearing heavy on his mother’s face, he said.
Studying artificial intelligence for the last three years, Jia, an assistant professor of business analytics and information systems at Quinnipiac University, wanted to provide additional support for his family and others who are experiencing grief. Artificial intelligence, he said, is a way to allow them to unload their emotions at any time of the day without straining others.
Jia found that many of the current chatbots did not include features that embodied human values including trust, well-being and calmness. Some stored sensitive personal information.
Seeking to change the norm, Jia recently developed an ethical framework for designing AI chatbots that he hopes will support people experiencing grief related to chronic illnesses. At the core of the framework is a focus on human values.
“Chronic illness often brings a form of grief that extends far beyond a medical diagnosis,” Jia said. “AI has the potential to provide meaningful support, but only if it is designed with human values and ethical safeguards at its core.”
Collaborating with researchers from the University of Florida and National Central University in Taiwan, Jia conducted a study of AI chatbot used for managing grief among those with chronic illness, with “Alzheimer’s disease as a specific case study.” The study found significant gaps in the current approaches to support people with grief, he said.
Jia said there are challenges ahead for developing an ethical framework for AI chatbots, which include issues related to privacy, safety and lack of long-term evidence on the effectiveness of the new technology.
“AI can give wrong or harmful answers,” he said.
But he said his framework, including the human values of calmness, informed consent, privacy, courtesy, human welfare, trust and autonomy in the design of AI chatbots, are a step in the right direction.
He said the idea is to provide extra support for those who are grieving, not to replace interactions with humans. And he hopes it can help caregivers.
“I just want to give them some space to take a breath,” he said. “Sometimes, if they have some emotional feelings, especially at grave moments, you can turn to the AI chatbot. I think the main benefit of using the AI chatbot is that you can talk with it 24/7.”
More people are turning to AI chatbots for mental health support. A study in JAMA Network Open found that one in eight young adults and adolescents used AI chatbots for mental health advice.
In Connecticut, psychologists said they are seeing numerous clients consult AI chatbots for help and that there are significant concerns about the technology and its safety.
Some early studies are showing that people using AI chatbots designed with cognitive behavior therapy principles experienced decreases in symptoms of anxiety and depression, according to psychologists. But psychologists say more studies are needed to understand how effective AI chatbots are in helping people with mental health needs.
“In general, there’s concern that in more extreme cases for youth who are using it in acute distress, those youth would receive advice to withhold information about the degree of their distress,” said Amber Childs, professor of psychiatry at the Yale School of Medicine and a clinical psychologist. “There have been at least two documented cases that we’ve seen of youth dying by suicide in connection to companion or AI generative chatbots.”
Childs said there are “ongoing concerns about AI chatbots providing youth with misinformation, misguidance, and non-evidence-based solutions that can bring cumulative harm.”
She said consultation with an AI chatbot may also deter the person from getting the mental health resources they may need.
In June, Gov. Ned Lamont signed legislation requiring AI chatbot operators to “make reasonable efforts to detect suicidal ideations or indicators when users express self-harm,” and to respond with appropriate resources, according to the governor’s office.
Dr. Benjamin Tolchin, director of the Yale New Haven Health Center for Clinical Ethics, said he is not aware of evidence that AI-based interventions for people who are grieving are actually effective at reducing symptoms of depression. He said there is still a need for more studies.
He said there are concerns that constant interactions with AI chatbots can replace interactions with other people.
“This may not be a great substitution for talking to other family members, a human therapist or friends,” he said.
Last fall, Brown University released a study on AI chatbots, finding that the chatbots, “even when prompted to use evidence-based psychotherapy techniques, systematically violate ethical standards of practice established by organizations like the American Psychological Association,” according to a release on the study.
The study found that chatbots are prone to ethical violations, including “inappropriately navigating crisis situations, providing misleading responses that reinforce users’ negative beliefs about themselves and others, and creating a false sense of empathy with users.”
U.S. Sens. Chris Murphy and Richard Blumenthal joined Republican senators Josh Hawley, R-Mo., and Katie Britt, R-Ala., in introducing legislation last year that would ban “AI companions for minors, mandate AI chatbots disclose its non-human status and create new crimes for companions who make AI for minors that solicits or produces sexual content,” according to a release from Murphy’s office. In April, the Senate Judiciary Committee unanimously passed the legislation.
In developing his framework, Jia and his research team evaluated grief-support chatbots, reviewed 85 peer-reviewed studies and consulted geriatric psychiatrists.
Jia said the new framework can ethically support someone experiencing grief in offering “a calm place to talk, regular check-ins and links to professional or community support.
“The chatbot should not diagnose, promise recovery, push someone to ‘move on’ or handle a crisis by itself,” he said. “Its role is to add support between human care contacts, while serious, sensitive or high-risk needs need to go to trained people.”
He said the confidentiality of the AI chatbot “needs both strong technology and clear rules.
“A person’s conversations should not be used to train the AI without separate, clear permission,” he said.
Tolchin said there are theoretical concerns about some current AI chatbots for grief, which simulate a deceased person.
“I’m not aware of any evidence that those are helpful,” he said. “It may potentially interfere with coming to terms with the loss of the person. If you’re able to have these ongoing conversations with a simulation of the person, that may make it harder to recognize the finality of the loss.”
Jia said his framework still needs testing with patients and caregivers in real settings.
“Good design can reduce risk, but it cannot remove all risk,” he said.
Overall, Childs said, “generally there is no replacement for human connection.
“The opportunity for AI is around, ushering us back to the basics of human connection, serving as a conduit and an amplifier to getting us back to that,” she said. “But there is and can never be a replacement.”
Copyright 2026 Hartford Courant. All rights reserved. The use of any content on this website for the purpose of training artificial intelligence systems, algorithms, machine learning models, text and data mining, or similar use is strictly prohibited without explicit written consent.