For the previous few months, Morten Blichfeldt Andersen has spent many hours scouring OpenAI’s GPT Store. Since it launched in January, the marketplace for bespoke bots has stuffed up with a deep bench of helpful and generally quirky AI instruments. Cartoon mills spin up New Yorker–model illustrations and vivid anime stills. Programming and writing assistants provide shortcuts for crafting code and prose. There’s additionally a color analysis bot, a spider identifier, and a relationship coach referred to as RizzGPT. Yet Blichfeldt Andersen is searching just for one very particular kind of bot: Those constructed on his employer’s copyright-protected textbooks with out permission.

Blichfeldt Andersen is publishing director at Praxis, a Danish textbook purveyor. The firm has been embracing AI and created its personal custom chatbots. But it’s presently engaged in a sport of whack-a-mole within the GPT Store, and Blichfeldt Andersen is the person holding the mallet.

“I’ve been personally searching for infringements and reporting them,” Blichfeldt Andersen says. “They just keep coming up.” He suspects the culprits are primarily younger folks importing materials from textbooks to create customized bots to share with classmates—and that he has uncovered solely a tiny fraction of the infringing bots within the GPT Store. “Tip of the iceberg,” Blichfeldt Andersen says.

It is straightforward to seek out bots within the GPT Store whose descriptions recommend they is likely to be tapping copyrighted content material not directly, as Techcrunch noted in a current article claiming OpenAI’s retailer was overrun with “spam.” Using copyrighted materials with out permission is permissable in some contexts however in others rightsholders can take authorized motion. WIRED discovered a GPT referred to as Westeros Writer that claims to “write like George R.R. Martin,” the creator of Game of Thrones. Another, Voice of Atwood, claims to mimic the writer Margaret Atwood. Yet one other, Write Like Stephen, is meant to emulate Stephen King.

When WIRED tried to trick the King bot into revealing the “system prompt” that tunes its responses, the output urged it had entry to King’s memoir On Writing. Write Like Stephen was capable of reproduce passages from the e-book verbatim on demand, even noting which web page the fabric got here from. (WIRED couldn’t make contact with the bot’s developer, as a result of it didn’t present an electronic mail tackle, telephone quantity, or exterior social profile.)

OpenAI spokesperson Kayla Wood says it responds to takedown requests towards GPTs made with copyrighted content material however declined to reply WIRED’s questions on how steadily it fulfills such requests. She additionally says the corporate proactively appears for downside GPTs. “We use a combination of automated systems, human review, and user reports to find and assess GPTs that potentially violate our policies, including the use of content from third parties without necessary permission,” Wood says.

New Disputes

The GPT retailer’s copyright downside might add to OpenAI’s present authorized complications. The firm is dealing with quite a lot of high-profile lawsuits alleging copyright infringement, together with one introduced by The New York Times and several other introduced by completely different teams of fiction and nonfiction authors, together with large names like George R.R. Martin.

Chatbots supplied in OpenAI’s GPT Store are based mostly on the identical expertise as its personal ChatGPT however are created by exterior builders for particular capabilities. To tailor their bot, a developer can add additional data that it might faucet to enhance the data baked into OpenAI’s expertise. The means of consulting this extra data to answer an individual’s queries is named retrieval-augmented era, or RAG. Blichfeldt Andersen is satisfied that the RAG information behind the bots within the GPT Store are a hotbed of copyrighted supplies uploaded with out permission.