AI Monitoring at The New York Times

Unionized tech employees at The New York Times are escalating a dispute over the use of artificial intelligence tools to monitor their performance. The workers, represented by the NewsGuild of New York's Tech Guild, have filed grievances and an unfair labor practice charge, alleging that the company is using two internal AI systems—DX and Glean—to track and evaluate their work in ways that violate their collective bargaining agreement. The union claims these tools have been used to establish de facto performance quotas and that management has failed to provide necessary information about its AI plans, as required by labor law.

Background on the Dispute

The Tech Guild, which represents about 700 engineers, designers, product managers, and data analysts, argues that the use of DX and Glean amounts to deploying surveillance technology against workers. DX is an engineering productivity tool that claims to help companies measure output, generative AI use, and efficiency. Initially introduced internally as a way to improve the developer experience, the union says the tool has increasingly been used to benchmark individual performance. Ben Harnett, a software engineer and chair of the unit's generative AI committee, told The Verge that disciplinary situations have involved managers citing DX data, such as the number of pull requests per week, comparing them to industry standards. Harnett expressed concern that such metrics flatten the complexity of engineering work and can be used unfairly in performance reviews or disciplinary actions.

The Role of Glean

Glean, another AI tool used at the Times, is designed to help employees search internal knowledge bases, including wikis, GitHub documents, Google Docs, and emails. While its stated purpose is to improve productivity, union members worry that it can also be used to monitor workers. Harnett noted that if an employee works on a draft document or leaves a comment in a file, a manager could query Glean about their individual performance. The Tech Guild also suspects that recent disciplinary notices were generated using Glean, given their style and format. Additionally, Harnett pointed out that Glean can generate false information, leading users on "wild goose chases."

Union Actions and Contract Issues

The union believes the use of DX and Glean violates multiple parts of their contract, including provisions on privacy, monitoring, job descriptions, and the requirement to notify and bargain with employees. In addition to the Tech Guild's grievances, the Times Guild—which represents about 1,500 editorial, ad sales, and support staff—has also filed an unfair labor practice charge against the company. Both unions allege that Times management has refused to respond to their requests for information about current and future AI use, which they say violates federal labor law.

The Times Guild is currently negotiating a new contract and is pushing for strong AI protections. These include requirements that any AI tool used in journalism must involve human oversight, that AI-generated content be transparently labeled, and that staff be compensated if the company enters into AI training deals. The Times has used AI for specific reporting tasks, such as analyzing millions of documents related to Jeffrey Epstein or scanning satellite images of Gaza, but the union wants safeguards to prevent misuse.

Broader Industry Context

The dispute at The New York Times is part of a larger trend in the media industry. Journalists and tech workers at other outlets are also grappling with how AI should be deployed. In April 2026, unionized employees at ProPublica staged a 24-hour walkout, with AI use being a key issue. At McClatchy, which publishes newspapers like the Miami Herald and The Sacramento Bee, some staff withheld their bylines in protest of a generative AI tool that produces multiple versions of stories. These examples show that workers across the industry are demanding a say in how AI is implemented, particularly when it affects their jobs and the quality of work.

Harnett emphasized that the Tech Guild is not opposed to using AI altogether. Instead, they want to ensure that workers have a voice in how AI tools are adopted and used. Metrics like token usage or AI interaction frequency can create pressure to produce more, but they may not align with doing quality work. Harnett argued that such surveillance distracts from actually doing a good job, which should be the company's goal.

The New York Times spokesperson Danielle Rhoades Ha said in a statement that the company disagrees with the characterizations made in the grievances and that it will respond through normal contractual processes. She also noted that the company has responded to over 80 other information requests from the Guild in recent years and will respond to this one in due course.

The conflict highlights a critical tension in the modern workplace: the use of advanced technologies to monitor performance versus the rights of workers to privacy, fair evaluation, and collective bargaining. As AI tools become more pervasive, similar disputes are likely to arise in other industries. The outcome at The New York Times could set an important precedent for how unions can negotiate over the use of AI in the workplace.

For now, the Tech Guild is preparing for a potential legal battle, while also continuing to raise awareness about the issues. The broader media industry is watching closely, as the decisions made at the Times may shape the future of AI deployment in newsrooms and beyond. The union's efforts reflect a growing recognition that workers must have a seat at the table when it comes to technologies that can fundamentally alter their work lives.

Historical context shows that similar fights have occurred with earlier technologies. For instance, the introduction of computer monitoring in the 1990s led to union grievances about privacy and fair evaluation. Today's AI tools offer far more granular data, enabling employers to track everything from keystrokes to collaboration patterns. This raises new questions about what constitutes reasonable monitoring and how metrics should be used in performance reviews.

The Tech Guild has also pointed out that the use of AI for surveillance can have unintended consequences. For example, employees might alter their behavior to game the metrics, which could actually harm productivity and innovation. Harnett noted that when people feel watched, they tend to avoid taking risks or trying new approaches, which are essential for creative and technical work. This suggests that the company's approach may be counterproductive in the long run.

Meanwhile, the Times' use of AI in journalism presents its own challenges. While AI can help analyze large datasets, it also raises ethical questions about bias, accuracy, and transparency. The Times Guild's push for human oversight and labeling reflects these concerns. Other news organizations have faced similar issues. For example, CNET faced backlash in 2023 after it was revealed that the site had been publishing AI-generated articles with little human oversight, leading to numerous errors. That incident highlighted the need for careful governance of AI in news production.

As the dispute at The New York Times unfolds, it is clear that the intersection of AI, labor rights, and journalism will remain a contentious area. Workers are increasingly organized and vocal about their concerns, and companies must navigate these demands while also seeking to leverage new technologies. The outcome of this particular fight could influence how other media companies approach AI adoption in the years to come.

In the meantime, the Tech Guild continues to gather evidence and build its case. The union has called on supporters to pay attention to the issue and to show solidarity with workers who are pushing back against what they see as invasive monitoring. Whether through bargaining, legal action, or public pressure, the union aims to ensure that the use of AI at The New York Times is conducted in a way that respects workers' rights and promotes a healthy, productive work environment.

Source: The Verge News