December 9, 2024 |

Newmark Journalism School Teach Communities to Find Truth Behind Algorithms

by Anna Leah

lowers the dance music she plays during her morning class at Craig Newmark Graduate School of Journalism at 勛圖弝け. Her students had been coding natural language models for the transcripts Vo collected of the ���Demure��� TikTok videos that went viral over the summer.

One of the investigative data journalist���s skills is knowing what language to use in different situations. With her young students, she gets their attention with memes. For one of her current projects, about misinformation and internet algorithms, Vo used the simple Vietnamese she learned at home to teach community elders how to spot deep-fakes.

The group had never heard of artificial intelligence before. Vo explained to an Oakland, California, group the differences in AI-generated images and real ones. ���Hair is not normal,��� she said, pointing to the computer-generated picture.

���Researchers were so fascinated with the problem, they forgot to bring people a solution,��� says Vo about misinformation online. Her approach is two-pronged: she investigates the algorithms ���wreaking havoc��� on communities, then teaches them how to find better sources with workshops like she ran that day in Oakland.

Vo works in data journalism, the fairly new branch of the field in which journalists analyze large data sets ��� or create their own ��� to conduct investigations and tell stories. Data journalists have found racially disproportionate patterns in IRS audits and created maps from EPA toxic emissions databases. To explore misinformation and social media suggestions, Vo manually coded Vietnamese transcripts of YouTube videos, which were generated by the site���s language detection software.

���Any sort of automated decision-making processes are AI,��� Vo explains about algorithmic suggestions online. ���Whatever comes up to your recommended list on YouTube��� is powered by millions of data points that are then turned into a model that then guess what you might like.���

In, Vo found Vietnamese immigrants in the Bay Area were left without timely in-language local news. As the pandemic increased, so did the need for information as well as a rise in anti-Asian hate crimes that caused community members to isolate further. These factors left the people vulnerable to misinformation from a vlogger, pushed into their feeds from YouTube���s algorithms.

One of Vo���s subjects received a third of their news from YouTube. Vo found this by analyzing 1,000 randomly sampled videos from four years of viewing history. An Information Futures Lab fellowship at Brown University funded her research.

At Buzzfeed, Vo taught readers and in what Facebook algorithms deliver to a conservative mother and liberal daughter. She has presented her work about algorithms and misinformation at Harvard, MIT, and Georgetown.

Through, Vo has been training journalists since May.

���These technologies that have been developed by the West, but are being exported at a fast speed towards the rest of the world, have not been developed with the Global South in mind,��� says Vo. She has educated journalists on four continents and wants them to understand the technology as well as potential local, labor, and environmental impacts. Vo says it���s important for journalists to have, ���the tools, the vocabulary, and the ability to hold these technologies accountable as they come into their countries.���

Vo joined Newmark 10 years ago. In the past year, she became an associate professor and began reporting for, a nonprofit outlet that covers immigration for an immigrant audience.

Recently, Vo to step away from breaking news and editorial pressures. have supported her journey.

���I actually had more time to think about what is the best way of going about this, because I was anchored in an institution that valued my ideas like 勛圖弝け,��� says Vo. ���I could expand my thinking and go away from how a journalist would approach it, but how would someone who researches this topic approach it?���

leads Vo���s department at Newmark. Junnarkar helped develop The New York Times��� digital presence and teaches while also working part-time at Bloomberg. In 2019, Junnarkar from the Ethics and Governance in AI Initiative, a joint project of the MIT Media Lab and Harvard���s Berkman Klein Center for Internet & Society. Newmark was the only journalism school selected.

In collaboration with Newmark���s professional development arm and Center for Community Media, the educators used the grant to train community media – journalists who work primarily for hyperlocal audiences or who serve specific groups. The workshops trained reporters to report on how AI used by government agencies and corporations can negatively impact immigrants and low-income communities.

For Vo, some of these questions began at home. She started questioning her own digital footprint in 2016, but expanded that view after the election. ���People started asking themselves, ���What happened to an information ecosystem?������ Vo remembers. She was inspired to explore ���What kind of information was everyone consuming that, quite frankly, a lot of the media organizations were not consuming.���