Pacific Islands, explained.
A leap of faith in Muani, Kadavu, Fiji (Carolyn Van Houten/The Washington Post via Getty Images)
As Pacific elections loom, Fiji’s disinformation crisis shows why AI-era democracy needs community verification networks.
Subscribe to The Informer for monthly expert analysis, and to Events for advance notice of visiting world leaders and distinguished guests.
You may unsubscribe from Lowy Institute newsletters at any time. For information on our privacy practices and how to unsubscribe, see our Privacy Policy.
Who fact-checks the coconut wireless?Who fact-checks the coconut wireless?
The Pacific is entering a concentrated stretch of electoral activity. New Caledonia held provincial elections on 28 June (Opens in new window). The Cook Islands votes on 12 August (Opens in new window). Fiji’s prime minister has confirmed the country’s next election (Opens in new window) will fall between 24 December 2026 and the first week of February 2027, and Papua New Guinea follows by mid-2027 (Opens in new window). This stretch comes at a moment when disinformation has never been more rife or more accessible, and when generative AI is introducing threats the region’s democratic infrastructure was not built to meet.
Fiji shows how vulnerable Pacific democratic institutions already were to disinformation, long before generative AI arrived to exploit that vulnerability. That history began during the Bainimarama government era, when between 2010 and 2023 Fiji’s Media Industry Development Decree (Opens in new window) vastly narrowed what could be reported on government and the judiciary, under penalty of fines and imprisonment. The effect was immediate displacement: local audiences who could no longer rely on mainstream reporting were redirected towards social media, blogs and word of mouth (“coconut wireless”), channels without editorial accountability. By the time the decree was repealed, a generation had learnt to treat unverified information as a normal substitute for verified reporting.
With few credible domestic news sources left to compete for trust, Facebook became the dominant source of news. Generative AI extends this dynamic further: anyone with an internet connection can now produce content in iTaukei, Fiji Hindi, or local dialects that reads as authentically Fijian, at a volume no Pacific newsroom or platform moderation team can match.
Pacific democracies are only as resilient as the information environment they depend on.
The consequences of disinformation are already established: rumours on social media escalating (Opens in new window) into roadblocks, school closures, and vigilante violence documented across the region. AI-enabled impersonation is a step further. In July 2025, a deepfake video (Opens in new window) falsely depicted Fiji Television’s news manager, Stella Taoi, endorsing a fraudulent investment scheme. Fiji TV called it “a serious breach of personal privacy and professional integrity”.
The weaponisation of a trusted journalist’s face to deceive the public illustrates the dangerous disinformation environment documented across the Pacific. Such deception erodes public trust in the institutional foundations democratic participation depends on.
In the months preceding this election cycle, accounts have begun to appear across Pacific Facebook feeds presenting as credible local commentators, contributing to political discourse at a volume and consistency that may indicate coordination. While these have yet to be formally documented, the Pacific Islands Forum Secretariat and Pacific Fusion Centre’s 2025 Pacific Security Outlook Report (Opens in new window) points to the trajectory: it expects the proliferation of mis- and disinformation to intensify ahead of national elections and during legislative processes, and warns that as deepfakes and data weaponisation advance, distinguishing authentic information from fabricated content will become progressively harder to do.
Acknowledging the risk of disinformation to democratic processes, in December 2025 the Fijian Elections Office and the Online Safety Commission signed a three-year memorandum of understanding (Opens in new window) to monitor election-related online harm and coordinate a response. That mandate already extends to AI-generated content, with the Stella Taoi incident prompting Online Safety Commissioner Filipe Batiwale to confirm as much (Opens in new window). However, what the MOU cannot do is reach the WhatsApp groups and hyper-localised newsfeeds where much disinformation actually circulates.
A generation had learnt to treat unverified information as a normal substitute for verified reporting (UNclimatechange/Flickr)
Beyond monitoring, the Pacific Islands Forum Secretariat has called for a strong, independent, and well-trained media to combat disinformation, reflecting an instinct toward top-down remedies. At the community level, improving media and information literacy (MIL) has been the default goal, but achieving MIL improvements in the Pacific has proven challenging.
Experience with MIL programs in Africa (Opens in new window) helps explain why they often fall short in the Pacific. Most MIL programs in developing countries have reached journalists and city-based professionals, while the health workers, religious leaders, teachers, and market vendors who carry information through communities – often via the “coconut wireless” – have rarely received systematic support. This is precisely the gap a Pacific approach should avoid repeating.
Treating media and information literacy as democratic infrastructure means building a direct line between the people Pacific communities already trust and the people equipped to verify what they are seeing. This matters most for what institutions are worst placed to catch. Institutional verification works on checkable facts: a date, a record, a source. A bot account built and maintained over months offers none of those – only a community member who knows the terrain well enough can notice something is off.
Establishing that capacity requires training community contacts, drawn from churches, women’s groups, and village councils, in a basic protocol: how to flag a suspicious claim, whom to send it to, and what to tell their community while a check is underway. In Fiji, their reports could feed directly into the monitoring unit established under the MOU between the Elections Office and the Online Safety Commission, giving it visibility into village-level rumour it cannot currently reach alone.
A similar model could resonate across the Pacific and does not rely on slow-moving platform-level moderation in minority languages, or regional treaty processes.
Pacific democracies are only as resilient as the information environment they depend on. Fiji’s experience points to what is required for the Pacific: trained community verification networks operating at the speed disinformation does.
About the authors
Michaela Long
Michaela Long is a Pacific specialist working across investment strategy, trade, and regional policy.
Connor Graham
Connor Graham is a Research Fellow in the Pacific Islands Program at the Lowy Institute.
The most-pressing world events explained by Lowy Institute experts and global contributors, in your inbox, every Thursday.
You may unsubscribe from The Interpreter at any time. For information on our privacy practices and how to unsubscribe, see our Privacy Policy.
