Dhrumil Mehta
@dmil.bsky.social
Data Journalism Prof. @ Columbia University and Deputy Director of Tow Center for Digital Journalism.
Great work by @aisvarya17.bsky.social and @klaudia.bsky.social. Read more below!
www.cjr.org/tow_center/w...
www.cjr.org/tow_center/w...
Why AI models are bad at verifying photos
Many of the latest large language models, including OpenAI’s recently released GPT-5, tout an impressive capability: reasoning about images. We have seen viral trends of AI models geolocating obscure ...
www.cjr.org
August 26, 2025 at 10:25 PM
Great work by @aisvarya17.bsky.social and @klaudia.bsky.social. Read more below!
www.cjr.org/tow_center/w...
www.cjr.org/tow_center/w...
Yes please! Just started BlueSky-ing. Thanks!
March 6, 2025 at 11:36 PM
Yes please! Just started BlueSky-ing. Thanks!
This has implications for news publishers who are considering making deals with AI companies. More on that in the piece. www.cjr.org/tow_center/w...
We Compared Eight AI Search Engines. They’re All Bad at Citing News.
AI search tools are rapidly gaining in popularity, with nearly one in four Americans now saying they have used AI in place of traditional search engines. These tools derive their value from crawling t...
www.cjr.org
March 6, 2025 at 11:34 PM
This has implications for news publishers who are considering making deals with AI companies. More on that in the piece. www.cjr.org/tow_center/w...
It turns out this problem extends to many other chatbots too. In @towcenter.bsky.social's tests, they often cited the wrong article, and sometimes also fabricated links & cited syndicated or plagiarized articles instead of the original.
March 6, 2025 at 11:34 PM
It turns out this problem extends to many other chatbots too. In @towcenter.bsky.social's tests, they often cited the wrong article, and sometimes also fabricated links & cited syndicated or plagiarized articles instead of the original.
This builds on prior tests showing that ChatGPT frequently cited the wrong source article, but rarely expressed uncertainty. If it didn't know the answer (at least to the kinds of questions we asked), it confidently made one up. cjr.org/tow_center/how-chatgpt-misrepresents-publisher-content.php
March 6, 2025 at 11:34 PM
This builds on prior tests showing that ChatGPT frequently cited the wrong source article, but rarely expressed uncertainty. If it didn't know the answer (at least to the kinds of questions we asked), it confidently made one up. cjr.org/tow_center/how-chatgpt-misrepresents-publisher-content.php