Software designers test specialized Chatbots fed by AI This can give medical advice and diagnose conditions – but questions abound on precision.
This spring, Google unveiled an “AI AIGERNIED” functionality where the company’s chatbot responses began to appear above typical search results, including for health related requests. Although it could seem like a good idea in theory, there were problems with the health advice offered by the software.
During the first week that the bot was online, a user said Google Ai gave incorrectPerhaps deadly information on what to do if it is bitten by a bell snake. Another research has led Google to recommend to people to eat “at least one small rock per day” for vitamins and minerals – advice withdrawn from a satirical article.
Google says they have since limited the inclusion of satirical and humor sites in their glimps and have deleted some of the research results that have become viral.
CBS Saturday morning
“The vast majority of AI previews provide high quality information, with links to dig more deeply on the web,” a Google spokesperson told CBS News. “For health requests, we have always had a strong quality and security railings, including non-responsibility clauses that remind people that it is important to ask for expert advice. We have continued To be referred when and how we show IA previews to ensure that the information is high quality and reliable.
CBS News confirmed that these fixes had not prevented any disinformation on health. Requests on the introduction of solid foods to infants under six months old have returned advice at the end of June. Babies are not supposed to start eating solid foods before the age of at least six months, according to the American Academy of Pediatrics.
Research on the benefits for the health of questionable trends in well-being such as detoxes or Drink raw milk Also includes demystified complaints.
Despite the outright quirks and errors, many health leaders say they remain optimistic about AI chatbots and how they can change industry.
“People will access the information they need,” said Dr. Nigam Shah, Data Chief Scientist at Stanford Healthcare. “In the short term, I am a little pessimistic, I think we are getting a little ahead of ourselves. But in the long term, I think these technologies will do us a lot of good.”
Other defenders of chatbots are quick to emphasize that doctors do not always do things. Estimates vary, but a Study of the Ministry of Health and Social Services From 2022, he found that 2% of patients who go to the emergency room each year can undergo damage after a erroneous diagnosis of a health care provider.
Shah compared the use of chatbots in the first days of Google himself.
“When Google Search has arrived, people were panicking that people diagnose themselves and all hell is unleashed. This has not happened,” said Shah. “Same thing. We will go through this phase (where) the new ones which are not entirely formed will make mistakes, and some of them will be bad, but overall, have information when there has no other option a good thing.
The World Health Organization is one of the companies that plunge its toes into AI waters. The organization’s chatbot, Sarah, draws information from the WHO site and its trusted partners, making the responses less prone to factual errors. When asked how to limit the risk of a heart attack, Sarah gave information about stress management, sleep well and focus on a healthy lifestyle.
Continuous progress in design and monitoring could continue to improve these robots.
But if you turn to an AI chatbot for health advice today, note the warning that comes with the Google version: “The quality of the information may vary.”
