The Guardian and the fake articles created by ChatGPT
Chris Moran, Chief of Editorial Innovation at The Guardian, revealed in an opinion column published in the newspaper that over the past few months, they have received inquiries about articles cited by ChatGPT that had never been written.
“Last month one of our journalists received an interesting email. A researcher had come across mention of a Guardian article, written by the journalist on a specific subject from a few years before. But the piece was proving elusive on our website and in search”.
The inquiry even made the quoted journalist doubt the veracity of the title, but they eventually concluded that ChatGPT had invented the citation.
There is a double-edged problem when it comes to artificial intelligence and journalism: humans are inclined to believe everything machines tell us, and on the other hand, artificial intelligence does not recognize its own ignorance. Essentially, it doesn’t know that it doesn’t know. Responding with “I don’t know” is not in its repertoire of options, and perhaps this makes it more similar to humans than we usually believe.
The article also highlights the complexity and opacity of generative artificial intelligence (AI) and how companies are rapidly implementing this technology to gain competitive advantages and satisfy shareholders. However, for responsible news organizations, the urgent question is how this technology can benefit responsible information in a time when misinformation, polarization, and bad actors are putting pressure on the information ecosystem.
Parallelly, The Guardian has established a task force and a small engineering team to learn about Artificial Intelligence, consider public policy, intellectual property questions, listen to academics and professionals, engage with other organizations, consult and train their staff, and safely and responsibly explore how the technology works when applied to journalistic use. As they progress in this process, they have discovered that they are contemplating potential uses in journalism and, above all, what makes it valuable.
While they have not yet announced a new format or product based on generative AI, The Guardian will publish a clear and concise explanation in the coming weeks on how they plan to employ this technology. The organization is committed to maintaining the highest journalistic standards and remaining accountable to their readers and the world for the journalism they publish.
While it is true that the digital ecosystem has been shaken in recent months by the emergence of generative artificial intelligence, little has changed in how we should approach such phenomena: ethics, critical thinking, and verification.
More info:
::: ChatGPT is making up fake Guardian articles. Here’s how we’re responding :::
