Developers aren’t really happy with their posts being fed into ChatGPT

• For developers, Stack Overflow is an important resource.
• Over the years, Q&A platforms have helped people navigate the ups and downs of programming.
• Concerns are currently arising after OpenAI signed a deal to use data to train ChatGPT.

Most developers have paid attention to Stack Overflow at some point in their career.

Since 2008, this site has served as a near-essential Q&A resource for engineers looking to find solutions to programming problems, improve their working knowledge, and connect with others in the software business. I did.

According to Stack Overflow, the site asks a new question on average every 14 seconds and has tallied approximately 60 million questions and answers to date. The company was acquired by European investment group Prosus in 2021 for $1.8 billion.

This makes it a very valuable resource, and it’s natural that developers will protect it aggressively.

It’s perhaps no surprise that some Stack Overflow users have started an uproar after the wealth of information they’ve contributed to the site over the years has been targeted by the data-hungry company OpenAI.

OpenAI contract sparks controversy

On Monday, the ChatGPT maker and Stack Overflow announced a partnership to provide OpenAI users and their customers with “the accurate, vetted data foundation that AI tools need to solve problems.”

OpenAI brings “verified technical knowledge from Stack Overflow directly to the surface on ChatGPT,” allowing users to access “trusted, attributed, accurate, and advanced technical knowledge developed approximately 16 years ago.” “It gives you easy access to code backed by millions of developers.” site.

For OpenAI, this deal is a no-brainer.

AI models like GPT-4 greatly benefit from being trained on as much data as possible. If trained on highly technical and specialized data like the one found on Stack Overflow, the model will respond better when responding to his ChatGPT user’s prompts.

Stack Overflow also hopes to benefit from the partnership by using OpenAI’s models in the development of OverflowAI. The product, announced in July 2023, was the company’s attempt to integrate generative AI capabilities into its services.

But some developers dedicated to Stack Overflow are starting to vent their frustrations.

On Mastodon, an open source social media service, one Stack Overflow user shared that he tried to delete the “top rated answers” on the site to protest the contract with OpenAI.

“Stack Overflow doesn’t allow you to delete a question that has an accepted answer and a lot of upvotes, because doing so would mean losing knowledge from the community,” wrote a user named Ben. “So we changed the top-rated answer to a protest message instead.”

Within an hour, the user said his attempt to change his top-rated answer was revoked, and his account was then suspended for seven days.

A screenshot of the Stack Overflow response shared by Ben says, “You recently deleted or altered content from your posts. When you post questions or answers on this site, those posts may be posted by other users. Be part of the collaboration ” also contributed to that content. ”

Ben went on to say that Mastodon is “a reminder that anything you post on any of these platforms can and will be used for profit.” “It’s only a matter of time before all messages on Discord, Twitter, etc. are deleted,” he suggested. They are scraped off, incorporated into models, and sold. ”

Meanwhile, on the Stack Overflow user forum, another user based in Europe asked, “Where can I find an opt-out option to prevent my answers from being used by OpenAI?” However, it also raised the question of whether European Union data privacy rules would allow responses to be removed from the site.

Users also expressed their opinions on the contract with Company X.

Emily Bender, a professor at the University of Washington, criticized the partnership Thursday, writing: After all, their principles were for sale. ”

Gergely Orosz, author of The Pragmatic Engineer newsletter, wrote in a post on Wednesday: “How do you react as a developer when you realize that your efforts to solve other developers’ problems (by answering questions on StackOverflow) are now useless? OpenAI improves ChatGPT performance Is there a way for StackOverflow to sell this data in order to

Stack Overflow did not respond to requests for comment.

While it’s clear that people are upset about this deal, it’s worth noting that this isn’t the first time OpenAI’s use of data created by other companies has raised concerns.

The company is facing multiple lawsuits from creators, including artists and writers, alleging that OpenAI is using their works without permission for profit.

Now that Stack Overflow’s integration with OpenAI is public, developers should consider how comfortable they are with ChatGPT by leveraging their insights.