Traders Buy New York Times Call Options Amid Unsealed OpenAI Lawsuit Filings

CNBC identified a purchase of roughly 4,400 New York Times call spreads amid newly unsealed filings in the publisher’s copyright case against OpenAI and Microsoft. The internal statements raise questions about whether AI answers compete with journalism, while the purchase reveals nothing certain about the buyer’s motive.

What the filings brought into view

The Times filed its suit in December 2023 and says it is owed billions of dollars, without naming a precise amount. The material was unsealed as U.S. District Judge Sidney H. Stein considered requests to decide parts of the case without a trial.

One question is what happens when a chatbot supplies an answer instead of sending a reader to the reporting behind it. CNBC, citing The Wall Street Journal, says an OpenAI executive allegedly called chatbots an “existential threat” to journalism because readers could get information without clicking through.

The filings also describe OpenAI staff discussing ways around publishers’ paywalls. CNBC reports allegations that OpenAI copied millions of copyrighted articles for training and may have used workarounds to retrieve them. A separate internal warning, dating to 2020, said AI could substitute for the labor of people who shape culture. The discussions and allegations are not court findings.

A short-dated trade, not a court timetable

The reported purchase paired October calls at $72.50 and $77.50, costing $0.925 per share. Such a spread can benefit from a rise in the stock within a defined price range. CNBC said Times shares had been flat for the year and barely moved on the Friday the filings drew attention, even as options activity increased.

CNBC’s Michael Khouw read the October expiry as a possible expectation of a settlement or ruling within weeks—not evidence that either is scheduled. He also noted trading in lower-priced put options and suggested the call buyer may have sold puts to cut the cost. The volume does not establish that one trader made both transactions.

the largest theft of labor in human history

Microsoft researcher Brent Hecht, as reported by CNBC citing The Washington Post

The fair-use question remains open

The reported employee statements give the Times material to cite against the companies’ fair-use defense, which argues that some uses of copyrighted work are permitted without permission. But an employee’s characterization does not decide whether the training was lawful or whether AI answers compete with the publisher’s work.

In a September 1 filing, the Justice Department asked the court to hold that training AI models on copyrighted works can qualify as fair use, citing national security. Its position does not resolve the Times’ claims about how articles were obtained or whether AI answers compete with the publisher’s work.

Khouw argues that the disclosures increase the chances of a large settlement and licensing deal or a Times victory at trial. That is his forecast. Neither the filings nor the options purchase settles what the judge will decide.