One Nation, the Australian party led by Pauline Hanson, has shown an AI-generated campaign video that The Guardian calls a collection of racist tropes. Guardian Australia political reporter Josh Butler argues that Australian watchdogs currently have almost no power to act against the clip, and that cheap AI tools are filling politics with what he calls “ensloppification”.
What happened
The Guardian reports that the video was reportedly played at One Nation’s Victorian campaign launch over the weekend. It shows a man who appears to be African holding a machete, an apparently Middle Eastern man wearing an explosive vest, a South Asian man in a turban talking about “mass migration”, and an Indigenous elder talking about people receiving money “on the basis of need, not identity”. The first part of the clip, which drew most of the media attention, is an AI-generated scene of a young man “coming out” as a conservative to his leftwing family. The progressive campaign group GetUp pointed out that this scene copies a similar Israeli campaign video from Benjamin Netanyahu.
Butler writes that most coverage of the launch barely mentioned the video. Reporters focused instead on Hanson’s reception, the music, and a pledge by candidate Mark Nicholson to jail Daniel Andrews and Jacinta Allan. Guardian Australia has previously been barred from One Nation events and did not try to attend.
The Guardian argues that the more important problem is enforcement. The Australian Communications and Media Authority cannot act, because its powers do not cover this kind of content and the video so far appears to exist only online. Electoral commissions can only regulate misinformation about the electoral process itself, not lies or offensive material. News Corp reported a One Nation spokesperson saying a broadcast version was being prepared. That could technically bring it within the authority’s remit, but one industry source told The Guardian the ad is unlikely to breach current broadcasting rules.
Pressure for change has been building. Senator David Pocock has spent two years pushing for rules on generative AI in election campaigns, and 35,000 people have signed his petition asking Labor to ban deepfakes in elections before the next federal vote. According to The Guardian, the federal Labor government has so far brushed off these calls. In a submission to parliament’s joint committee on AI, the Australian Electoral Commission warned that deepfakes circulating during voting “may well compromise the actual or perceived integrity of an election.” Professor Terry Flew of the Centre for AI Trust and Governance told the inquiry that the government should consider banning AI deepfakes during election periods.
The One Nation clip is not the only example. The Guardian also points to an AI ad from the conservative group Advance showing Daniel Andrews popping champagne corks after a Labor win. AAP fact-checked Liberal MP Jason Wood over a fake image of Anthony Albanese at a deserted airport, and has debunked several AI-generated interview-style videos supposedly filmed outside Parliament House.
Why it matters if you live in United Kingdom
The story is about Australian law and an Australian state election, and The Guardian’s analysis does not discuss British regulation. Readers in the UK should therefore not assume that the gaps it describes, or any proposed fixes, apply here. The wider argument still applies outside Australia.
Butler’s main point is that AI makes this kind of content much easier to produce. Making such a video used to mean hiring a film crew and finding actors willing to perform the material, which was harder and more expensive. Today it can be generated cheaply. A political video showing staged scenes, recognisable figures or stereotyped groups no longer tells you how much money or organisation sits behind it.
The article also separates two kinds of harm. Butler accepts that no voter will believe the leftwing family in the One Nation clip is real. The concern is about tone and targeting rather than deception. Professor Flew told the inquiry that AI-generated deepfakes “go beyond the legitimate bounds of public campaigning” and can be “genuinely threatening to the individuals and groups who are the targets of them.” Many internationals in Britain belong to communities often shown in such material, and this point speaks directly to them.
The Guardian also notes that this content is designed as rage bait. It aims to provoke a reaction and then present that reaction as proof the party is being “picked on”. Its approach is to report on how the content works without boosting it, and the same approach suits anyone deciding whether to share or comment on similar videos in their own feeds.
What to watch next
Three things deserve attention. The first is whether One Nation releases the broadcast version its spokesperson described, and whether the Australian Communications and Media Authority looks at it. The second is the parliamentary joint committee on AI, where witnesses told members this week that action was overdue. The third is whether the federal government, which The Guardian says is working on legislation covering a range of AI harms this year, decides to include election deepfakes after earlier resisting calls to do so.
These concerns are not new. A parliamentary report from October 2024 had already warned that AI could be used “to adversely influence or undermine electoral and democratic processes.” The Guardian’s argument is that the problem is now visible in real campaign launches. Whether Australia changes the law in response will give other democracies, including the UK, a useful example to follow.
Source: The Guardian
Sources
- The Guardian: One Nation’s AI video is vile and tired – but the ensloppification of Australian politics doesn’t stop there, accessed 8 October 2026
The UK, for internationals
One email a week: the week's news for internationals in the UK, new practical guides and what changed in the cities. No spam, unsubscribe any time.
