Stories
Slash Boxes
Comments

SoylentNews is people

Submission Preview

No link to story available

NYT disputes OpenAI “hacking” claim by pointing to ChatGPT bypassing paywalls

Accepted submission by Freeman at 2024-03-12 18:43:44 from the AI overlords dept.
News

https://arstechnica.com/tech-policy/2024/03/nyt-disputes-openai-hacking-claim-by-pointing-to-chatgpt-bypassing-paywalls/ [arstechnica.com]

Late Monday, The New York Times responded [arstechnica.net] to OpenAI's claims that the newspaper "hacked" ChatGPT to "set up" a lawsuit [arstechnica.com] against the leading AI company.
[...]
OpenAI had argued [arstechnica.com] that NYT allegedly made "tens of thousands of attempts to generate" supposedly "highly anomalous results" showing that ChatGPT would produce excerpts of NYT articles.
[...]
But while defending tactics used to prompt ChatGPT to spout memorized training data—including more than 100 NYT articles—NYT pointed to ChatGPT users who have frequently used the tool to generate entire articles to bypass paywalls.

According to the filing, NYT today has no idea how many of its articles were used to train GPT-3 and OpenAI's subsequent AI models, or which specific articles were used, because OpenAI has "not publicly disclosed the makeup of the datasets used to train" its AI models. Rather than setting up a lawsuit, NYT was prompting ChatGPT to discover evidence in attempts to track the full extent of copyright infringement of the tool, NYT argued.
[...]
"In OpenAI’s telling, The Times engaged in wrongdoing by detecting OpenAI’s theft of The Times’s own copyrighted content," NYT's court filing said. "OpenAI’s true grievance is not about how The Times conducted its investigation, but instead what that investigation exposed: that Defendants built their products by copying The Times’s content on an unprecedented scale—a fact that OpenAI does not, and cannot, dispute."
[...]
On an OpenAI community page [openai.com], one paid ChatGPT user complained that OpenAI is "working against the paid users of ChatGPT Plus. This time they’re taking away Browsing, because it reads the content of a site that the user asks for? Please, that’s what I pay for Plus for."

"I know it’s no use complaining, because OpenAI is going to increasingly 'castrate' ChatGPT 4," the ChatGPT user continued, "but there’s my rant."

NYT argued that public reports of users turning to ChatGPT to bypass paywalls "contradict OpenAI’s contention that its products have not been used to serve up paywall-protected content, underscoring the need for discovery" in the lawsuit, rather than dismissal.
[...]
Unlike book authors, who alleged that ChatGPT removed copyright management information (CMI) without providing examples of infringing outputs, the NYT provided receipts, NYT argued. And while book authors "focused their claim on how the removal of CMI could induce third parties to infringe," NYT is arguing that "removal of CMI 'facilitates' or 'conceals' OpenAI’s own infringement."

NYT wants a court to not only award damages for profits lost due to ChatGPT's alleged infringement, but also to order a permanent injunction to stop ChatGPT from infringement. A win for NYT could mean that OpenAI could be forced to wipe ChatGPT [arstechnica.com] and start over. That could perhaps spur OpenAI to build a new AI model based on licensed content—since OpenAI said earlier this year it would be "impossible" to create useful AI models without copyrighted content [arstechnica.com]—which would ensure publishers like NYT always get paid for training data.

Previously on SoylentNews:
OpenAI Says New York Times 'Hacked' ChatGPT to Build Copyright Lawsuit [soylentnews.org] - 20240301
Why the New York Times Might Win its Copyright Lawsuit Against OpenAI [soylentnews.org] - 20240220
New York Times Sues Microsoft, ChatGPT Maker OpenAI Over Copyright Infringement [soylentnews.org] - 20231228
Report: Potential NYT lawsuit could force OpenAI to wipe ChatGPT and start over [soylentnews.org] - 20230821

Related stories on SoylentNews:
Microsoft in Deal With Semafor to Create News Stories With Aid of AI Chatbot [soylentnews.org] - 20240206
AI Threatens to Crush News Organizations. Lawmakers Signal Change Is Ahead [soylentnews.org] - 20240112
Writers and Publishers Face an Existential Threat From AI: Time to Embrace the True Fans Model - 20230415 [soylentnews.org]


Original Submission