The ban comes as healthcare providers rush to embrace generative AI to diagnose, read images and suggest treatment plans.
By RFA Cantonese
Authorities in central China have banned doctors working for online clinics from using AI to prescribe medication to patients, according to several media reports.
“Online healthcare providers must be connected to the provincial medical insurance electronic prescription center and circulate electronic prescriptions in accordance with regulations,” health authorities in Hunan province said in a statement dated Feb. 25.
“It is strictly prohibited to use artificial intelligence and other automated methods to generate prescriptions,” it said.
The ban comes as healthcare providers around the country rush to embrace DeepSeek and other homegrown AI tools, potentially eroding trust in an already corrupt healthcare system.
Under current regulations, doctors in China are allowed to use AI tools to aid them in their work, including reading medical images, organizing data and formulating treatment plans, AI Hospital Operation and Management Research Institute director Wei Zining told the Chengdu Business Daily newspaper on Feb. 23.
“It’s hard to say how things will be in 100 years’ time, but [for now], AI is only allowed to assist doctors, not replace them,” Wei said.
Doctors working for online prescribing services must also undergo real-name authentication before seeing patients, to ensure that the medical services are provided by them alone, and are banned from using human or AI substitutes, the paper said.
State media have published a slew of recent reports about healthcare providers who use China’s newly emerging homegrown generative AI tools, including DeepSeek, in their treatment of patients.
The Fuyang People’s Hospital in the eastern province of Anhui announced on Feb. 21 it would be using DeepSeek to “analyze cases, discover potential patterns of disease, and to assist in optimizing diagnosis and treatment plans,” according to a report published on the Anhui provincial government website.
‘Consultation guidance system’
The model is also being used to carry out medical triage and signposting, the article said.
“The DeepSeek consultation guidance system can understand the symptoms described by patients through natural language processing technology, recommend appropriate departments and doctors, and provide the best appointment options based on doctor schedules and patient time preferences,” it said.
The ban comes as state media outlets including state broadcaster CCTV have been quick to laud the use of AI in medical settings.
“Patients only need to open our hospital’s WeChat official account and ask in voice or text, ‘What should I do if I have stomach pain?’ or ‘Which department should I go to for a headache?,’ and AI can quickly give thoughtful advice and medical guidance,” it quoted Fu Qihua, deputy director of Sichuan Provincial People’s Hospital as saying.
“Some automated tasks, like intelligent medical guidance, drug distribution and other repetitive and mechanical nursing tasks, are being replaced by automated systems,” Kang Dan, a nurse at the Huaihua No. 2 People’s Hospital, told the station.
“Be we also need to be particularly vigilant about issues such as nursing ethics, nursing data security and patient privacy protection,” Kang said.
China’s recent advances in AI and big data, including its recently launched DeepSeek AI model, will also likely boost the government’s surveillance capabilities, given its widespread access to personal and private data on its citizens, according to Feb. 11 report from the Washington-based National Endowment for Democracy.
Translated with additional reporting by Luisetta Mudie. Edited by Malcolm Foster.
“Copyright © 1998-2023, RFA.
Used with the permission of Radio Free Asia,
2025 M St. NW, Suite 300, Washington, D.C. 20036.
https://www.rfa.org.”
