complain, complain, complain, complain, complain, complain, complain, complain. In Spanish, there are at least a dozen ways to tell someone they have the flu — it depends on the country.
It is also difficult to translate “cardiac arrest” into Spanish because “to arrest” means being taken into custody by the police. Similarly, “Drunk” This means you have food poisoning, it doesn’t mean you’re drunk.
There are countless examples of translation errors in any language: words take on new meanings, idioms come and go, and communities adopt slang and dialects for everyday life.
Human translators work hard to keep up with changes, but California soon plans to hand that responsibility over to technology.
State health policy officials want to use emerging artificial intelligence technology to translate a large portion of documents and websites related to “health and social services information, programs, benefits and services.” as per state recordsSami Gallegos, a spokesman for California’s Health and Human Services Agency, declined to say what documents and language would be included, saying the information is confidential.
The agency is inviting bids for the ambitious initiative, though its timeline and cost are not yet clear. Gallegos said human editors overseeing the project will oversee and edit the translations.
Agency officials said they hope to save money and make important health care forms, applications, websites and other information accessible to more people in the nation’s most linguistically diverse state.
The project will begin by translating written material. Agency Secretary Mark Ghaly He said that if this technology is successful, it can be implemented more widely.
“How do we transform not only all of our documents, but also our websites, our ability to have conversations, even some of our call center inputs around AI?” Ghali asked during an interview. April Briefing on AI in Healthcare in Sacramento.
But some translators and scholars fear the technology lacks the subtlety of human interaction and is not up to the challenge. They say handing this sensitive work over to machines could lead to errors in wording and understanding – ultimately making information less accurate and less accessible to patients.
“AI cannot replace human compassion, empathy and transparency, meaningful gestures and tone,” he said. Rithy LimA medical and legal interpreter working in Fresno for 30 years, she specializes in the Cambodian and Khmer languages.
artificial intelligence is the science of designing computers that simulate human thinking. A type of artificial intelligence known as Generative AIOr GenAI, in which computers are trained using massive amounts of data to “learn” the meaning of things and respond to signals, is making a big difference. A wave of investment led by companies like Open AI and Google.
AI is increasingly being integrated into healthcare, including Programs to Diagnose Diabetic Retinopathy, Analyze mammograms And Connecting patients with nurses remotelyTechnology promoters often claim that soon everyone will have their own “AI Doctor,
AI has been a game changer in the field of translation as well. ChatGPT, Google’s Neural Machine Translation And open source Not only are they faster than older technologies like Google Translate, they can process vast amounts of content and nearly mimic human translation using a huge database of words.
Where it might take a professional human translator three hours to translate a 1,600-word document, AI can do it in a minute.
Arjun “Raj” ManraiAssistant Professor of Biomedical Informatics at Harvard Medical School and Deputy Editor of the New England Journal of Medicine AI.said the use of AI technology represents a natural progression in medical translation, given that patients already use Google Translate and AI platforms to translate for themselves and their loved ones.
“Patients are not waiting,” he said.
Generative AI could be particularly useful in this context, he said.
He said these translations “can provide real benefits to patients by simplifying complex medical information and making it more accessible.”
In its bid documents, the state said the project aims to “increase the speed, efficiency, and consistency of translation and improve language access,” as 1 in 3 people in the state speak a language other than English and more than 200 languages are spoken.
In May 2023, the State Health and Human Services Agency will “Language Access Policy“which required departments to translate all “critical” documents into at least five major languages spoken by Californians who have a limited command of English. At the time, these languages were Spanish, Chinese, Tagalog, Vietnamese, and Korean.
Examples of important documents include applications for state programs, notices regarding eligibility for benefits, and public website content.
Currently, human translators create these translations. With AI, more documents can be translated into more languages.
A survey conducted A survey conducted late last year by the California Health Care Foundation found that 30% of Spanish speakers have difficulty explaining their health problems and concerns to a doctor, compared to 16% of English speakers.
Health equity advocates say AI will help close this gap.
“This technology is a very powerful tool in the field of language access,” said Sandra R. Hernandez“In good hands, there are many opportunities to expand translational capacity to address inequities,” said Dr. , President and CEO of the Foundation.
But Hernandez cautioned that AI translations must be humanly monitored to ensure the correct meaning is understood.
“Human interaction is so important to make sure you get accuracy and reflect cultural nuances,” he said.
Lim recalled an incident in which a patient’s daughter read the pre-operative instructions to her mother the night before surgery. Instead of translating the instructions as “You can’t eat anything after a certain time,” she told her mother, “You must not eat anything.”
Mother had eaten breakfast, and the surgery had to be rescheduled.
“Even a few words that change the meaning can have a huge impact on the way people consume information,” he said. Sejin PekDoctoral Candidate in Digital Journalism, Human-Computer Interaction, and Emerging Media at Boston University.
Paik, who grew up speaking Korean, also pointed out that AI models are often trained From a Western perspectiveHe added that the data that drives translation filters languages through an English perspective, “which can result in misinterpretations of the second language.” Amid this rapidly changing landscape, “we need more diverse voices, more people thinking about ethical concepts, how we can best anticipate the impact of this technology.”
Manrai pointed to other flaws in this nascent technology that must be addressed. For example, AI sometimes invents sentences or phrases that are not in the original text, thereby potentially creating false information – a phenomenon that AI does not understand. Scientists call this “hallucinations” Or “conversation.”
Ching Vong, executive director of the Vietnamese Community Health Promotion Project at UC San Francisco, has been translating health materials from English to Vietnamese and Chinese for 30 years.
He gave examples of language nuances that can confuse AI translation programs. For example, breast cancer is called “chest cancer” in Chinese, he said.
And “you” has different meanings in Vietnamese, depending on the person’s ranking in the family and community. If a doctor uses “you” incorrectly with a patient, it can be disrespectful, Vuong said.
But Gailey stressed that the opportunities outweigh the drawbacks. He said the state must “foster innovation” to help vulnerable populations get greater access to care and resources.
And he stated clearly: “We will not replace humans.”
