The AI ​​battle is raging within the New York Times

How newsrooms should use AI — or whether they should — has been a constant debate in the media industry for the past few years. Increasingly, these rules are being discussed at the negotiating table between organizations and publishers. Currently, employees at The New York Times they are preparing for a fight.

Tech Guild employees say Time management refused to provide the agreement with information about the company’s use of AI, its plans to use AI in the future, and how it will affect operations and operations. (The union filed an unfair dismissal claim earlier this month.) The Tech Guild, New York’s NewsGuild organization of about 700 software developers, designers, product and project managers, and data analysts, also filed a complaint. Time Managers broke their collective bargaining agreement when they started using two internal AI tools that track and monitor employee performance and activities.

One of the AI ​​tools, called DX, advertises itself as an engineering tool that allows companies to track employee productivity, AI usage, and efficiency, among other metrics. DX was announced internally as a way to improve the productivity of developers, says Ben Harnett, a software developer at the company. Time and chair of the unit’s generative AI committee. Purpose, at least according to Time management, was testing the entire company. In the past few months, DX data has become more popular, with more individual branding, Harnett says.

“Now disciplined people are suddenly reading to them, ‘You only did (pull requests) once a week, for everything, and it’s 25 percent below the industry standard,'” says Harnett. He is concerned that blanket metrics control all of the work that team members do and reduce engineering to abstract metrics that can be used against employees to discipline or evaluate performance. The metrics don’t correlate to the quality of work or the actual amount of product an employee provides, says Harnett.

“All of that (information) can be expected to … help us understand how we’re doing, but not how they’re using it and tracking it, which we think is limited,” Harnett said. Seaside. DX figures have been mentioned in recent discussions, the Tech Guild says.

Another program called Glean takes internal knowledge bases such as wikis, GitHub documents, Google Docs, and emails, and allows employees to search the system to find the information they need easily. But there is concern among employees that Glean can also be used to manage employees because they pull a lot of internal documents: Harnett says that if they are working on writing documents to explain what they are building or leave comments in a file that is available in Glean, for example, the manager can ask the tool about their actions or contributions. The Tech Guild said Seaside that the style and format of the latest information sent to employees indicates that it was created using Glean. Harnett says Glean has problems – it creates false positives and can lead the user to “goosebumps.”

“The way they’re using (DX and Glean) we feel like they’re using the latest employee evaluation methods,” says Harnett. The union believes that the use of these tools violates several sections of its contract, including privacy protection and supervision, job descriptions, and requirements for employee notification and consultation.

Both the Tech Guild and the Times Guild (which represent 1,500 writers, brands, and contributors at Time) filed a lawsuit against unfair labor practices Timeclaiming that the company violated labor laws by refusing to respond to their request for information about the use of AI in the retail environment.

The Time He did not respond to specific questions about DX’s use of Glean, but spokeswoman Danielle Rhoades Ha said in an email that the company does not agree with the allegations in the complaint and will respond as part of a “general agreement.”

“In the same way, we will respond to the Request for Information (RFI) as soon as we have done with the other 80+ RFIs from the Guild in recent years,” said Rhoades Ha.

The Times Guild is currently negotiating a new agreement, pushing for stronger protections against AI, such as requirements that a person is behind every AI tool being used, that any journalism that uses AI be clearly registered, and that workers pay for training on AI models that the company can create. The Time It uses smart tools to generate other reports, such as using them release millions of documents related to Jeffrey Epstein or download satellite images of Gaza trying to find where Israel dropped a certain type of bomb.

Journalists across the industry are in the midst of discussing social networking, and AI is one of the most pressing issues at stake. In April, 150 employees joined ProPublica he stopped working for 24 hours; One of the main sticking points with the management was how AI could be used and revealed to the audience. After McClatchy, the company that publishes newspapers like Miami Herald and The Sacramento Beeit started coming out an AI creation tool that spits out different types of storiessome workers blocked their lines in protest.

Harnett emphasizes that the group’s position is not that AI should not be used, but that employees should have a say in how it is being used. Metrics such as the number of tokens an employee uses or how often they use AI to do their job create pressures and incentives that are not related to doing a good job.

“It’s distracting (you) from doing a good job, which is what we think the company should want,” he says.