ChatGPT’s “Browse” browser stirred up controversy recently when users realized it may give them access to paywalled content, leading OpenAI to temporarily disable the feature while they fix its loopholes. The news came after an alarming Reddit post revealed the feature had provided full access to a paywalled article from The Atlantic. It’s speculated that the technology might be able to ‘bypass’ payment implementation by using search engine optimization cached versions of pages.
OpenAI, the creator of ChatGPT, responded quickly to this issue and released a July 4th tweet relative to the disabling of its Browse feature. The post highlights how OpenAI is “doing right by content owners”, to prevent serious implications that follow data scraping for AI training. Similarly, for weeks, fear about data scraping stirred up a major scandal with Elon Musk citing it as the reason for new limits on tweets.
Ultimately, what this issue stands to show is the top priority of content owning companies to ensure proper payments and credit to creators at the end of the day, something which goes without saying for OpenAIt too, even ethrough its Beta testing phase. A class action lawsuit was filed back in June ―also connected to the usage of data scraping to train ChatGPT’s AI― drives forward accountability claims that are taken very seriously by AI developers.
Aside from unfavourable association, ChatGPL becomes more noteworthy, as showcased by recent edition of our Magazine (AI Eye: AI travel booking hilariously bad, 3 weird uses for ChatGPL, crypto plugins). Here it’ll be seen as a frontrunner in open sourcing tools and solutions with ethical conduct and assurance in payment, helping innovators stay updated and keep initiative.