Meet the AI leaders: Data reporter and AI specialist Cynthia Tu

ByAnabelle Nicoud

Two years into the generative AI boom, many newsrooms and publishers are rethinking their editorial roles to include an AI-focused mentality. The New York Times, Hearst,AFP, USA Today… many news organisations of all sizes have recently created new roles working in editorial or product, an illustration of the importance AI has rapidly acquired in media.

What do these heads of AI do?

In a series of interviews, Anabelle Nicoudtalked to leaders in the United States and Europe to find out.

In July 2023 Open AI announced a partnership with the American Journalism Project(AJP) to support local newsrooms. One of these projects is ongoing at the Sahan Journal, a nonprofit digital newsroom for immigrants and communities of colour in Minnesota.

We talked to Cynthia Tu, data reporter and AI specialist, who began her job in January 2024.

How did you start working in your current role?

I started as a data reporting fellow right after I graduated from grad school. Before that, I dipped my toes into AI, playing with ChatGPT. And we talked about generative AI in grad school. But I never really understood the entire scope or how it could apply to journalism.

A few months into my fellowship, we got a grant opportunity with AJP to do a two-year experiment and see how we could ethically apply AI in a newsroom. The proposal is to use AI technology to help with the business side of the newsroom. So instead of using it for editorial and reporting tasks, we want to use AI to help generate more revenue and open up new revenue streams.

You come from editorial, but you're working with AI on the business side. Why make that choice; what opportunities do you see?

It's important for me, and for our newsroom, to develop the skills to understand both the limitations and advantages of this technology so we can effectively use it in our day-to-day operations. At Sahan Journal, because we are a small local newsroom, it's crucial to free up space and resources to make workflows more efficient.

So far, we think AI is the best, most efficient, and cost-effective way to help us do that.

Can you tell me more about what you worked on?

Our first experiment is in the testing phase with our advertising project, helping our advertising associate manager learn more about our audience, and pitch that data to their clients.

Essentially, we have an audience insight report with tons of audience data. We thought it would be great to finally utilise this data, so I built a custom GPT using the audience report as a knowledge model, and the output provides descriptions of potential clients.

Do you see a lot of potential for Gen AI to be applied in the newsroom?

It definitely has a lot of potential for developing internal tools. To be honest, we are somewhat suspicious of AI technology because we report for immigrants and communities of colour, and we know that AI can have biases.

'Our editors and reporters are hesitant to fully trust AI to produce content for our audience, and I think that's a fair concern. So instead, we can use AI to assist with more mundane tasks that aren't reader-facing.'

AI can be a helpful assistant that frees up our hands to focus on meaningful, creative work, like aggregating information and gaining insights from large data sets.

Data literacy is key in the newsroom. How do you train your newsroom, and did it shift perceptions a bit?

We started with a training session for the editorial team by demystifying what generative AI is. Interestingly, one of the reporters said, 'Oh, I just use it as another search engine,' but that's not really what ChatGPT is built to do. It's actually a language model, not a logical model.The training also sparked a discussion about creating AI ethics guidelines for the newsroom and publishing them on our website, so people know what we're using AI for and why we're using it.

What do you see a year from now in your use of AI tools?

I see a lot of just learning what's new and trying to keep up, because it evolves so fast. There are many opportunities, but we're also facing a lot of challenges, which means we'll need to do a lot of education and ongoing training to prepare for that.

About the author, Anabelle Nicoud

A freelance journalist and consultant based in San Francisco, Nicoud currently collaborates with The Audiencers newsletter and the Canadian monthly L'actualité.

She has worked with Apple News+ (2022-2024); helped the editorial teams at La Presse (2015-2019) and Le Devoir (2019-2022) with their digital transformation, while leading ambitious editorial projects that have won prestigious journalism awards in Canada and Quebec.

A former journalist for La Presse and correspondent for Libération in Canada, Nicoud is passionate about the impact of technology on the media, she closely follows issues related to the use of artificial intelligence.

WAN-IFRA External Contributor

[email protected]