close
close

Stack Overflow could suspend your account if you change a post to protest OpenAI’s deal

Developer code laptop

Luis Alvarez/Getty Images

A deal between Stack Overflow and OpenAI appears to have sparked a battle between the developer forum and its users.

On Monday, Stack Overflow announced a new deal in which OpenAI would collect user content to train ChatGPT. As a forum for developers and programmers, Stack Overflow hosts technical posts and content valuable to a generative AI service like OpenAI’s ChatGPT.

However, the announcement forced at least one user to change their posts in protest. This, in turn, caused Stack Overflow to suspend this user’s account for a week. In a post on Mastodon, an Epic Games developer named Ben said that he attempted to remove content from his Stack Overflow posts in response to the forum’s partnership with OpenAI.

Also: Stack Overflow uses AI to give programmers new access to community knowledge

Since Stack Overflow doesn’t allow you to delete questions that have answers and lots of upvotes, Ben protested by changing the content of his questions to read: “I asked this question in protest of Stack Overflow’s decision to partner with OpenAI , removed. This move steals the work of everyone who has contributed to Stack Overflow without the ability to unsubscribe. OpenAI has a history of flooding the web with inaccurate information and explicitly states that they will never pay creators for their work.

The modification of a question by a Stack Overflow user

ben_ui/X

Within an hour, the forum moderators had reset Ben’s questions to their original state and suspended his account for seven days. Regarding Ben’s Mastodon post, a message from the Stack Overflow moderation team informed him that he had recently removed or redacted content from one of his posts.

“Please note that once you post a question or answer to this site, those posts become part of the collective efforts of others who also contributed to that content,” the statement continued. “Posts that may be useful to others should only be removed in exceptional circumstances.”

The moderators concluded by saying that once the matter is resolved, his reputation score will be restored and his account will continue to operate as usual.

Message from Stack Overflow from the moderation team

ben_ui/X

Following his first Mastodon post, Ben said that he asked Stack Overflow to permanently delete his questions and answers in accordance with GDPR. He also criticized OpenAI’s data scraping.

“It’s just a reminder that anything you post on any of these platforms can and will be used for profit,” Ben wrote. “It’s only a matter of time before all of your messages on Discord, Twitter, etc. are scraped, fed into a model and sold back to you.”

Other Stack Overflow users have reached out with questions and complaints about the OpenAI deal. One person asked, “Where is the opt-out option so that my answers are not used by OpenAI?” Another person asked whether Stack Overflow users are legally entitled to benefits from the OpenAI deal.”

The conflict raises the question: Who owns the data you post in a public forum? The GDPR provides for a “right to be forgotten,” where you can request that your data be removed from a website. However, this usually applies to personal or sensitive information.

Stack Overflow’s terms of service state: “Once you post content to the public space, you willingly give up some rights and control over that content.” Additionally, the forum allows you to delete a question or post, but only if no one else does responded to it. Once a post generates replies and even upvotes, removing it is strongly discouraged.

Also: Stack Overflow joins Reddit and Twitter in charging AI companies for training data

This conflict also highlights issues surrounding generative AI and data collection. What happens if a public website allows its content to be scraped by AI? Do the creators of this content have a say in this matter?

Stack Overflow is facing backlash in part because the company had previously resisted the lure of AI. In a late 2022 policy post, the site banned the use of ChatGPT and other generative AI tools when writing or rewriting content. Now, AI seems to be fine as long as it generates revenue for the company.