The launch of a new tool, which uses artificial intelligence to adjudicate on the credibility of news articles, Objection AI, offers a daunting vision of the future of reputation management.  Dystopian tools like these provide a powerful reminder of why robust and independent press regulation is so important.

‍

Objection AI, created by businessman Aron D’Souza, is a new service that allows individuals and corporations to submit a claim to an AI system on articles alleged to be “inaccurate”. The tool then analyses the article and purports to give a verdict on its truthfulness.

‍

On paper, this software might sound useful and capable of having a positive impact. The process, as Objection AI claims it to work, seems straightforward:

1. File a complaint to Objection, which has an AI system scan the article for selected reasons for the claim.
2. Then, a human investigator—whether they’re from the CIA, FBI or are a defence analyst—will overlook the AI’s findings.
3. If the human investigator deems the claim and findings acceptable, it will then be pushed to a tribunal of Large Language Models, or LLMs, that will review the entirety of the article and the author’s previous publishings.
4. Within 72 hours, the platform publishes an accuracy review of the article and puts the journalist onto a forum called the Honor Index, which measures journalists’ credibility based on the review.

‍

But AI is not a mature or neutral technology. It is affected by biases, hallucinations and errors.  Its output is dictated by the information it has access to, and the guardrails set by its owner and operator.

‍

As a result, a service for determining the truthfulness of articles which relies on AI and is run by a private organisation is vulnerable to producing false or unfair verdicts, and wrongly tarnishing the reputations of journalists.

‍

Additionally, Objection AI costs $2000 to use. This puts it out of reach for most people.  If a service like this catches on, the majority of ordinary people will be priced out of any justice it is capable of delivering.

‍

There is also criticism for its methods and the factors it takes into account. The platform has pledged to penalize journalists for using anonymous sources, a relatively normal practice that journalists have utilized for decades. This could potentially compromise whistleblowers. This sort of approach also raises the risk of AI being used to identify anonymous sources, which can be done by recognising the language they use, their profession or expertise, or other relevant contextual factors in any article.

‍

D’Souza has had numerous run-ins with the press in the past couple years. This raises questions about his motives: is this about improving journalistic standards or is it about helping billionaires to combat coverage they don’t like?

‍

“The fact that this is a pay-to-play kind of system … tells me that they are less concerned about providing helpful information for the general public and much more concerned with giving the already powerful a means to basically browbeat their journalistic opponents,” said Jane Kirtley, a lawyer and professor at the University of Minnesota, speaking to TechCrunch.

‍

The PJP is also concerned that this tool targets the individual journalist.  We believe that publishers should be held to account for low standards, not individual reporters, who are often put under significant pressure by executives and whose output is subject to editorial control.

‍

At its worst, this new software could be a real risk to our democracy and the free press. Allowing anyone to pay money to target a journalist’s entire reputation because of dissatisfaction with an article is dangerous to the liberties we are supposedly guaranteed.

‍

“Why would you believe that AI would necessarily give you more reliable information about the truth or fals[ity] of fact than a journalist who had researched and written the story? I mean, why would you just assume that? I wouldn’t assume that at all,” Kirtley said, speaking to TechCrunch.

‍

The good news is that legally or in regulatory terms, this website has no standing.  The verdicts it would deliver have as much credibility or value as if they had been scribbled on the back of a napkin. The platform cannot force journalists to participate in its findings.

‍

But it does provide an insight into a dystopian future in which AI is left to regulate the media, and only the wealthy can access it.

‍

This is why the people need to get involved and protect our free and accountable press. The public deserves to know the truth without fear of silence and compliance to those who provide it.

‍

The solution is independent regulation led by human beings, who are transparently appointed, independent of both the press and politicians, and who handle complaints according to publicly available criteria.  This is what Leveson proposed, and is the most effective way of upholding freedom and accountability for the press.