Big Tech platforms and a loose collection of fact-checking organisations are shining more light on dubious claims, but finding the right answer needs to be as easy as a Google search.
This year will go down as an unprecedented one when it comes to democracy. Nearly half of the world’s population will vote in national elections held in at least 64 countries this year, including India, the United States of America, Pakistan, Indonesia, Mexico, the United Kingdom, as well as the European Union.
Not all of those elections will be free and fair, many will be bitterly contested, and the rise of content automatically generated by artificial intelligence could supercharge attempts to manipulate voters’ decisions. That’s on top of the deluge of fake facts and manipulated photos circulating on everything from Taylor Swift’s private jet trips to the war in Gaza. Covid-19 is still the subject of a steady feed of medical misinformation.
Fake news is going viral at an accelerating rate on social media platforms. The answer, we are told, is to take nothing at face value, only rely on trusted sources of information, and do our own research into far- fetched-sounding claims. Certainly, I’ve always adhered to the “Sagan standard”, that saying from the late, great American science communicator Carl Sagan, who reminded that “extraordinary claims require extraordinary evidence”.
But that’s easier said than done when you are scrolling through your Facebook newsfeed and deciding whether to like or share a post that is already going viral. Where are you supposed to go to quickly check the veracity of what you are about to blast out to your own network of contacts? How do you avoid becoming part of the problem?
Big Tech’s efforts to crowdsource fact-checking
The social media platforms themselves should be doing the heavy lifting for you, shutting down fake news and dubious claims before they go viral. They have the legions of content moderators and the filtering technology to do the job. But they do a patchy job of that at best.
X, formerly Twitter, has largely disbanded its fact-checking efforts in favour of a crowd-sourced system called Community Notes, which allows X users to add context to potentially misleading posts. If enough registered Community Notes contributors “from different points of view” rate the post as helpful, the note will be publicly shown on a post.
It’s better than nothing but seems largely as though Elon Musk is washing his hands of the expense and hassle of policing his platform. Facebook and YouTube do better on specific issues, such as Covid-19, where content will be automatically flagged as potentially containing inaccurate information and offer links to official sources of public health information. But mis and disinformation come in all sorts of different forms on a bewildering spectrum of topics that automated systems can’t deal with.
A handful of independent organisations have sprung up to try to address the most dubious and potentially damaging claims. They include Snopes, Politifact, and Factcheck.org (and the associated SciCheck). The largest, English-based ones are US organisations and focus on the US news agenda, though fact-checking not-for-profit and for-profit organisations are appearing around the world, including Factly, the India-based fact-checking platform, and Africa Check. The Duke University Reporter’s Lab last year identified 417 fact-checking organisations, up from 11 in 2008.
Mainstream media organisations around the world supplement those efforts, though maintaining a dedicated fact-checking team is difficult for cash-strapped media outlets. Stuff’s The Whole Truth series, was the only New Zealand-based fact-checking project listed by the Reporter’s Lab, and ran a useful series looking at medical misinformation in 2022 as well as a fact-check of political claims leading into the 2020 election. But, it is no more. Meanwhile, Australia has five fact-checking organisations listed.
What we really need is an index of authoritative, independent fact-checks that’s easy to find on, say, the Google search engine where more than 90% of internet searches are conducted globally. Google is doing just that with its Fact Check Explorer, which lets you search for and explore fact checks conducted by accredited organisations Google works with.
Fact Check Explorer uses a mark-up feature Google makes available to publishers, which makes fact-checked content more visible to Google. I visited Fact Check Explorer in the wake of Tucker Carlson’s exclusive YouTube-streamed interview with Vladimir Putin, puzzled as I was at some of the claims the Russian president seemed to be making.
“Ukrainians still consider themselves Russians. What is happening is an element of a civil war,” were Putin’s translated words. Really? I understood that Ukraine is a sovereign nation and what is happening in the country is a result of Russia’s invasion of Ukrainian territory, rather than a civil war.
Fact Check Explorer provided some good context, indexing a fact check from Polygraph.info, run by the Washington DC-based broadcaster Voice of America. Google doesn’t verify the facts it indexes at Fact Check Explorer; it is relying on the accuracy of the pre-vetted organisations who flag information for indexing.
A list of trending fact-check topics appears on the Fact Check Explorer website. The fact-checking organisation is clearly labelled and there’s a link to the fact check article. A “true”, “false”, “inaccurate” or “missing context” tag is parsed from the indexed article and included in Fact Check Explorer, giving you a very quick conclusion on the veracity of the claim you are seeking more information on.
Google Fact Check Explorer is a valuable service, though it is invisible on the main Google search page - you need to know it exists to do a quick fact-check, though a Google search should surface the articles featured in the index. The bigger problem is that Fact Check Explorer has very little content relevant to New Zealand society, politics and current affairs.
The New Zealand-related fact checks are mainly carried out across the Tasman by the likes of news agencies AAP and AFP. Fact Check Explorer only really surfaces major trending issues that have attracted a lot of debate already, so the random quack medical treatment you see people discussing on X or Reddit threads may not have any entry on Fact Check Explorer.
Images are a particularly powerful form of potential misinformation and, in the era of generative AI, can be more easily manipulated than ever. Google is trialling a new feature of Fact Check Explorer that allows an image to be vetted to see if it has previously been fact-checked.
Google last year spent US$13.2 million on helping establish a new independently administered Global Fact Check Fund to help fund fact-checking agencies around the world that are tackling misinformation. It comes as the relationship between Google, Meta, and New Zealand news publishers faces a major test.
Parliament’s select committee for economic development, science and innovation will this week hear submissions on the Fair Digital News Bargaining Bill, a piece of draft legislation introduced by the last government and aimed at requiring Google, Meta and Microsoft to strike commercial deals with news publishers to compensate them for running snippets of their content on social media platforms and in search results.
Google has a deep interest in ensuring its search results aren’t rife with misinformation and fake “facts” and relies on media outlets and fact-checking agencies to do the heavy lifting. So, it will need to tread carefully in opposing efforts to strike commercial content deals with media outlets.
But as the biggest indexer of the world’s information, Google also has the power to supercharge fact- checking efforts, making them easily accessible to hundreds of millions of internet users. Fact Check Explorer is a start but needs to be fully integrated into the Google search experience and feature more fulsome results to start to turn the tide on mis and disinformation.
