Additionally, the San Francisco-based startup said it plans to launch a paid service for businesses called “ChatGPT Business” with additional data controls.
The move comes as ChatGPT and other AI chatbots have come under scrutiny for their handling of user data, which is largely used to improve or train these systems.
Last month, Italy banned ChatGPT over privacy concerns, but said OpenAI could resume operations if it met certain requirements, such as providing users with tools to prevent their data from being exploited. France and Spain have also begun investigations into the technology.
Mira Murati, OpenAI's chief technology officer, said the company complies with European privacy laws and is working closely with authorities to resolve the matter.
A representative of the company that owns ChatGPT also said that the "anonymity" feature did not originate from the ban in Italy but had been developed over the past several months to help users "take an active position" regarding data collection activities.
“The company will be moving towards prioritizing user privacy,” Murati said.
Accordingly, the new update on April 25 allows users to turn off the "Chat History & Training" option in the settings and export data.
Nicholas Turley, a product officer at OpenAI, likened the new feature to incognito mode on internet browsers. But he said the company would still retain conversations for 30 days to monitor for violations and abuse before permanently deleting them.
The ChatGPT Business subscription service, which is expected to be released in the coming months, will not use conversation data for AI training by default.
Software giant Microsoft, after investing billions of dollars in OpenAI, is offering ChatGPT to businesses.
According to Reuters
Source
Comment (0)