In recent days, a series of posts on social media have attracted hundreds of thousands of views about Google secretly accessing all of Gmail content to train the Gemini AI model.
Not only that, a report from Malwarebytes also stated that Google is using private data from Gmail for the purpose of training AI models, and accused some accounts with this option of being pre- potential by default.
This information has made many Gmail users worried and sparked controversy about privacy with personal data.
Regarding this content, on November 22 (local time), Google posted an official notice on Gmail's X account to deny it.
The company also stressed not changing anyone's settings.
" Gmail's Smart feature has been around for years. Google does not use users' Gmail content to train the Gemini AI model. If there are any changes to terms or policies, we are always transparent, Google said.
According to Google, the Smart feature in Gmail, which includes composition suggestions, automatic sorting and completion of thu, operates based on separate models, not related to the use of personal content for Gemini training.
This is also a feature that has existed for a long time, not a new setup or a secret change as many recent social media posts have described.
Although Google has made a clear denial, the incident reflects the fact that users are very sensitive to privacy issues as AI is thriving.
Rapidly spreading misunderstandings show the need for timely information and communication transparency from leading technology companies.
