Inside Reuters’ AI Playbook: Why Jane Barrett Says Journalism Can’t Afford to Sit This One Out

Artificial intelligence is rapidly transforming the media landscape, and no organization embodies this shift more thoughtfully than Reuters. As one of the world’s largest and most trusted news agencies, Reuters is not just adapting to AI—they’re actively shaping how it integrates with journalism. Leading this charge is Jane Barrett, Head of AI Strategy at Reuters, who brings a unique perspective forged from years of experience straddling the editorial and business sides of the company. Her approach balances innovation with the rigorous ethics and trust principles that journalism demands.

In this article, we delve into how Reuters has embraced AI, the challenges and opportunities it presents, and why Barrett believes the journalism industry cannot afford to sit on the sidelines as AI reshapes newsrooms worldwide.

From Reporter to AI Strategist: Jane Barrett’s Journey at Reuters

Jane Barrett’s career at Reuters began in 1999 as a reporter, and over more than two decades, she has held a variety of roles across newsrooms and business units, spanning multiple countries. This extensive experience gave her a deep understanding of the newsroom’s inner workings and the business imperatives behind news production.

After a decade of reporting, Barrett transitioned into business roles, including launching a financial video service within Reuters. This move allowed her to combine editorial insight with business strategy, sparking a passion for innovation within journalism. When generative AI emerged in late 2022 and early 2023, she recognized it as the most significant transformation the industry would face and positioned herself to lead Reuters’ AI integration efforts.

“I want to be on the forefront of that. I want to ride that wave and not be crushed by it.”

The Early Days of AI at Reuters: From “Wait and See” to “Lead and Shape”

Before generative AI exploded onto the scene, Reuters was already experimenting with machine learning, especially in areas like market data and sports reporting where structured data made automation feasible. They even acquired a company called PLX to accelerate processing of corporate earnings reports using AI.

So when tools like ChatGPT arrived, Reuters didn’t start from scratch. Instead, they turned up the dial on existing AI efforts. Barrett and her colleagues quickly formed a scrappy, cross-functional team—including data scientists from Thomson Reuters Labs—to experiment and produce proofs of concept. These early projects demonstrated AI’s practical relevance to journalistic workflows.

By mid-2023, armed with solid evidence of AI’s potential, they convinced Reuters’ leadership to integrate AI tools more broadly. What began as a small “scrum squad” evolved into a structured initiative to embed AI across editorial processes and business operations.

Embedding AI in the Newsroom: Three Core Strategies

Reuters’ AI integration centers on three key pillars:

1. Reduce: Automating routine, time-consuming tasks to free journalists for higher-value work.
2. Augment: Enhancing existing content by repurposing it into new formats, languages, and social media-friendly snippets to reach broader audiences.
3. Transform: Exploring how AI can fundamentally change news production and distribution, anticipating future industry shifts.

With roughly 2,600 journalists worldwide, the newsroom is vast and fast-paced. Barrett emphasizes that tools must integrate seamlessly into existing workflows to be adopted. For example, one AI tool uses generative AI to extract facts from press releases rapidly but always keeps a human journalist in the loop to verify and edit before publication.

“We made sure the tool fits within the same UI that our journalists already use — so it’s frictionless and fast.”

This human-in-the-loop approach is critical. Journalists rely on AI to assist, not replace, preserving editorial standards and trustworthiness.

Current AI Tools in Use at Reuters

Reuters has launched several AI-powered tools embedded in their newsroom systems, including:

• Fact Extraction from Press Releases: Speeds up the breaking news cycle for financial clients by automatically pulling key facts while allowing journalists to verify instantly.
• Leon CMS AI Integrations: Incorporates transcription, translation, headline generation, bullet point summaries, and a GPT-4-based assistant that suggests story improvements or cuts.
• LAMP Packaging Tool: Uses AI to help editorial teams rank stories, select matching visuals, and package content efficiently for clients and Reuters’ own platforms.

These tools illustrate Reuters’ focus on augmenting journalists’ work rather than automating entire stories. Barrett stresses the importance of measuring success at the use-case level before stitching together an end-to-end AI workflow, always ensuring human oversight at critical points.

The Reuters AI Suite: Empowering Media Customers

Beyond internal newsroom innovation, Reuters has launched the Reuters AI Suite—a set of AI tools designed for their business customers, including news agencies and sports media outlets. This suite helps clients streamline video content production by automating transcription into over 150 languages and translating into seven production languages.

Rather than focusing on generative video or imagery, the suite emphasizes “reversioning” content—transforming existing material into new formats or languages quickly and accurately. Barrett notes this is just the beginning, with plans to expand the suite as new solutions emerge.

Generative Video and Imagery: A Strict No-Go for Reuters Journalism

One of the most critical editorial lines Reuters draws is around generative video and imagery. Barrett explains that news visuals must be faithful representations of actual events. Creating synthetic images or videos—even to extend existing footage—risks breaking the fundamental trust between the camera’s lens and the audience.

“We do not use generative AI for any images or video. A news image or video has to be a representation of exactly what we have seen.”

This ethical boundary is non-negotiable and reflects Reuters’ commitment to reliability and integrity. While generative visuals may have roles in other content types, for journalism, the line is clear.

Trust, Verification, and the Challenge of AI-Generated Content

As AI-generated content proliferates, verifying authenticity becomes increasingly difficult. Reuters employs a dedicated visual verification team using open-source intelligence to confirm the validity of user-generated and social media content, especially during breaking news.

Barrett highlights ongoing industry conversations about labeling or watermarking AI-generated content to prove provenance. Reuters has experimented with cryptographic hashes on photos in conflict zones to verify authenticity but also encountered challenges balancing security and transparency.

These efforts underscore the complex trade-offs between speed, trust, and verification in the AI era.

Copyright, Compensation, and Ethical AI Use

Reuters deeply values its content, which represents a costly and sometimes dangerous journalistic investment. Barrett expresses concern about proposals to extend “fair use” exemptions to AI training datasets, which could undermine the value of original journalism.

Reuters has negotiated deals with companies like Meta to license content for AI training but has not, to date, partnered with OpenAI. Barrett and her team experiment with publicly available models while exploring options to develop proprietary language models tailored for accuracy, latency, and control.

“We can’t wait for all the philosophical conversations to be settled before moving. The pace of AI development is too fast.”

Environmental concerns around the computational costs of large AI models also factor into Reuters’ strategic thinking.

Introducing AI to the Newsroom: Culture, Training, and Transparency

Barrett emphasizes that successful AI adoption depends on bringing journalists along on the journey. Reuters has invested heavily in mandatory training courses to build AI literacy across the newsroom, including workshops on effective prompting and tool use.

Monthly town halls and open office hours foster ongoing dialogue, addressing fears around AI replacing journalists and empowering staff to experiment and innovate. Barrett shares a memorable moment in Tokyo where a journalist’s eyes lit up after discovering how AI could help solve a work challenge.

This combination of top-down support and grassroots experimentation creates a culture where AI is a collaborator, not a threat.

Facing the Future: AI’s Impact on Journalism and Public Trust

As AI-driven platforms become primary news sources for many audiences, Reuters sees its role as a source of “truth” more crucial than ever. Barrett warns of the dangers of misinformation and disintermediation, where audiences lose direct contact with original news sources, undermining trust.

Reuters is investing in its own digital platforms to provide accurate, unbiased news and exploring how AI can enhance personalized, multimodal experiences. Barrett argues that AI companies must partner with reputable media organizations to ground their outputs in trustworthy information.

“AI companies should be doing deals with media companies to ensure answers are grounded in good, trusted, accurate information.”

Conclusion: Journalism’s AI Future is Now

Jane Barrett’s insights reveal a newsroom actively embracing AI not as a shortcut, but as a powerful collaborator that can enhance journalistic quality, speed, and reach—without compromising integrity. Reuters’ approach is a masterclass in balancing innovation with ethics, transparency, and human oversight.

As AI continues to reshape how news is produced and consumed, the industry’s survival depends on organizations like Reuters leading the way with thoughtful strategies and clear boundaries. Barrett’s work reminds us that journalism must engage with AI head-on, shaping the technology to serve the public good, not be shaped by it.

For media professionals and news consumers alike, Reuters’ AI journey offers valuable lessons in trust, adaptation, and the future of news in an AI-driven world.