ChatGPT security issue: After user points out serious security flaw affecting chatbot that stores conversations in plain text, OpenAI releases desktop version update and fixes issue

Washington, July 4: The OpenAI ChatGPT macOS app had a security issue that allowed chats to be stored on the computer in text format. OpenAI chats were stored on the Apple computer as plain text, allowing anyone to access and read the conversation between the user and the AI ​​chatbot. This serious security issue posed a major threat if any malicious app or bad actor gained access to the users’ computers.

According to report By Edgewith this access, an application or person could read the conversations between the user and ChatGPT, including all the data contained in them. The report mentions that Threads user Pedro José Pereira Vieito shared a video of how the problem worked. In the video, the user opened a ChatGPT window and asked a few questions. On the other hand, the user opened a text file. Koo shutting down: Indian social media platform created as X rival to be shut down as Dailyhunt acquisition talks fail, co-founder Mayank Bidawatka reacts.

Pedro José Pereira Vieito – ChatGPT Security Issue Video, watch here:

Post by @pvieito

Thread View

As the user typed text, it appeared in the file as plain text, as shown in the video. The user was seen using a program called “ChatGPTStealer” that made the conversation between him and ChatGPT appear as plain text. The text loaded when the user clicked the “Load Chat…” button shortly after asking the chatbot a question.

According to the report, Edge contacted OpenAI about this security issue, and the company released an update saying that chats will now be encrypted. Developer ChatGPT said it was aware of the issue and had begun shipping a new version of the app to encrypt the conversation between users and the chatbot. Google Search Console is not providing updated data due to “latency issues” with search performance reports. The company says it will “work on fixing this issue soon.”

OpenAI spokeswoman Taya Christianson confirmed that the company is committed to providing users with a great experience while maintaining high security standards. With the new update the problem has been solved, and users will no longer see conversations in plain text.

(The above story first appeared on LatestLY on Jul 04, 2024 11:50 AM IST. For more news and updates on politics, world, sports, entertainment and lifestyle, log on to our website latestly.com.)