While journalists of all stripes are being displaced by the rise of artificial intelligence cannibalizing the written word, there’s more than one bit of “AI” out there that’s causing a stir among journalists.
In recent months, there’s been controversy, such as in this Facebook post, surrounding AI-generated or automatically seeded articles on Inquirer.net’s website featuring an author going by the name Antonio Inares. These articles are often political in nature and, on a closer reading, all positive press for particular politicians.
In journalists’ circles, the suspicion is that Antonio Inares is a pseudonym for seeded or AI-generated political articles — thus, the initials “AI.” The running joke is, “Did we ever meet him on a coverage?” He didn’t even have a LinkedIn or any social media account that would provide a trail of having done journalistic field work, they said.
Whether Antonio Inares is a real person or not raises several ethical issues in reporting, among them:
- The non-disclosure of the use of AI in any part of the writing and editing process.
- The non-disclosure if the too-positive write-ups were sponsored by the politicians and therefore didn’t go through the usual vetting process of both reporter and editor.
It is also important to discuss what happens when a newsroom isn’t responsibly disclosing its use of technologies like AI, or its guidelines for such, if any.
How does one perceive a journalistic institution?
Journalists, by and large, are held to higher ethical standards when it comes to disclosure and reporting. Ethical journalism precludes that journalists with integrity provide accurate, fair, and thorough information on matters of import to people seeking news.
As such, journalists take responsibility for the accuracy of their work and fix any failings that may appear.
Because their work is done in service of the public good, journalists are generally accountable and transparent, within reason, when it comes to the nature of their work and their reporting. They cite sources when necessary, but also withhold the identity of sources if the safety of a source might be compromised in the telling.
Because journalism work is often in the public sphere and up for scrutiny, the perception of conflict of interest, irresponsibility, and corruption can be a deal-breaker for news seekers looking for a reputable institution to support.
Responsible AI in the service of journalism
Because journalists are accountable for their work and any failings attributed to it, the use of AI should ideally be judiciously questioned by a journalist.
While my opinions on AI skew negatively when it comes to the idea it can do any original writing or reporting — I still feel it regurgitates words more than writes in any meaningful way — AI can be used responsibly to help journalists do their work.
For example, AI can help transcribe video interviews or audio recordings. With proper prompting, AI can help find important points in documents, summarize a document accordingly, or even properly tag, add, or propose metadata to improve search engine optimization on an article.
It can even help organize thoughts so that writers can write better, though I still believe AI cannot replace the act of actual writing.
Proper disclosure of AI use
Because journalists are held to a higher standard, a standard which AI can neither perceive nor be held accountable to, it is important that AI use is properly disclosed on articles that are aided by it.
Rappler, for its part, used AI to begin the creation of our PHVote candidate profile pages in 2022. Rappler also released its guidelines on AI use in 2023. Iloilo’s The Daily Guardian also released its own set of AI usage guidelines in June 2024 after a year of consultations and benchmarking.
News organizations around the world are seeing the value, peril, and responsibilities presented by AI use in the newsroom and are responding appropriately. For instance, the Associated Press has a page dedicated to its standards around generative AI use. In another case, Reporters Without Borders and 16 partner organizations also released the Paris Charter on AI and Journalism in November 2023,
Not properly disclosing AI use or other forms of automation, such as the seeding of press releases, would be ethically compromising. Worse, it is a disservice to readers seeking the news if an institution has no information available for an author or doesn’t disclose that he isn’t real.
Further still, a news organization that doesn’t disclose such advertising — what else would you call positive PR for a politician — can be seen as corrupt by savvy news seekers who see through the ruse.
Not properly disclosing such things erodes trust in the process of doing journalism and in the journalists of the institution caught in the middle.
The only right way to use AI, if it must be used, is to ensure people know it was used in the process. The only right way to post political fluff is to disclose whether or not your institution has been paid to seed it.
Anything else is a disservice to the reading public. – Rappler.com