Was a protracted learn? Making it easier…
What is the story
OpenAI has addressed a safety flaw inside its just lately launched ChatGPT macOS app, which saved consumer conversations in plain textual content on a PC.
This difficulty was found and demonstrated by a Threads consumer named Pedro Jose Pereira Vieito, who created an app that would entry and show these conversations instantly after they occurred.
The issue with storing chats in plain textual content is that they are often simply accessed by hackers or malicious apps, making all interactions with ChatGPT readable.
An replace to encrypt chats launched
Upon being alerted by The Verge concerning the safety difficulty, OpenAI promptly launched an replace to encrypt the chats.
“We’re conscious of this difficulty and have shipped a brand new model of the applying which encrypts these conversations,” mentioned OpenAI spokesperson Taya Christianson.
She additional emphasised that “We’re dedicated to offering a useful consumer expertise whereas sustaining our excessive safety requirements as our expertise evolves.”
Successfully securing consumer conversations
Following the replace, Vieito’s app was unable to entry the conversations, they usually have been not seen in plain textual content.
Nevertheless, until customers decide out, OpenAI should still assessment ChatGPT conversations for security and to coach its fashions.
This privilege doesn’t lengthen to unknown third events that achieve entry and know the place to look.
