French Polynesia

By the close of this century, sustainable tourism in French Polynesia — Te Ao Mā'ohi — will look very different from the overwater-bungalow postcard that still dominates the global imagination, but it will not disappear. The most plausible picture, calibrated to roughly 2.5 to 3 degrees Celsius of warming above pre-industrial levels and to continued French sovereignty under some upgraded autonomy arrangement, is a smaller, geographically redistributed, and culturally re-indigenized industry, anchored in what is now the world's largest marine protected area and shaped at every level by an ambivalent layer of artificial intelligence. Tahiti will still receive visitors. Bora Bora will still exist. But the gravitational center of the trade will have moved.

To understand why, one has to start with where the territory stands today. The five archipelagos welcomed roughly 327,000 international visitors in 2024, with about 264,000 of those staying overnight on land — the highest total in the country's history and up nearly seven percent on the year before. Tourism contributes about fourteen percent of GDP directly, closer to seventeen or eighteen percent when indirect effects are counted, and supports around 13,000 jobs, or eighteen percent of all employment. The territorial strategic plan known as Fāri'ira'a Manihini 2027 set a ceiling of 280,000 annual tourists, deliberately calibrated to roughly match the resident population of 278,000, and that ceiling is already essentially being reached. Whatever succeeds it will have to confront the awkward question of whether to lower, hold, or quietly relax the cap.

The climate trajectory will force that conversation regardless. Under current policies, the South Pacific can expect a sea-level rise of roughly half a meter to a meter by 2100, with low-likelihood but high-impact ice-sheet scenarios extending the upper bound considerably further. The five archipelagos will experience this very differently. The Society Islands — Tahiti, Mo'orea, Bora Bora, Huahine, Raiatea — are high volcanic islands whose land sits well above any plausible rise; their vulnerability lies in shallow lagoon reefs and the iconic but exposed bungalow infrastructure built barely a meter above the waterline. The Marquesas, steep and almost reef-less, are paradoxically the least climate-exposed archipelago and almost certainly the biggest relative growth zone of the coming decades. The Australs, cooler and home to one of the more advanced Indigenous-governed marine zones, will see modest but steady growth built around humpback whale-watching and low-density cultural tourism. The Gambier lagoons have so far weathered heat events better than the northern atolls and will likely remain the core of a smaller, more boutique pearl industry.

The Tuamotu atolls are the central, contested case. The seventy-eight atolls of the group typically rise only two to four meters above mean sea level, and yet geomorphological work on Pacific atolls has shown that healthy reef-fringed islands can actively accrete sediment and even gain land area as the sea rises — provided the reef continues to produce the coral rubble that feeds the shoreline. The brutal corollary is that where reefs collapse, that sediment factory shuts down. The most likely outcome is a partial habitability: the larger, well-managed atolls such as Rangiroa, Tikehau and Fakarava persist as niche dive and cultural destinations, with permanent infrastructure rebuilt on piles or relocated to the most stable motu, while smaller and lower-elevation atoll resorts consolidate or close. The reefs themselves, after successive bleaching events including the 2024 austral-summer episode that affected up to sixty percent of corals around the Society Islands, will most likely persist in a profoundly simplified form — lower cover, fewer species, more weedy and stress-tolerant — rather than disappear outright.

Against this backdrop, the single most consequential governance event of the decade occurred in Nice in June 2025, when President Moetai Brotherson announced the upgrade of Tainui Atea to the formal status of the world's largest marine protected area. Two fully protected zones totaling 900,000 square kilometers now sit alongside artisanal-fishing belts of roughly 186,000 square kilometers extending fifteen to thirty nautical miles around the archipelagos, with deep-sea mining and drifting fish aggregating devices banned across the entire five-million-square-kilometer exclusive economic zone. This is the marketing and moral anchor for the second half of the century. It transforms French Polynesia from a beach destination into an obligatory case study in global ocean governance, and it gives the territory a story to tell that is far more durable than coral cover in any single lagoon.

Artificial intelligence runs through every layer of this future as a deeply uneven productivity layer. The positive uses are already operational and will mature: the Coral Gardeners platform on Mo'orea, ReefOS, uses solar-powered cameras and cloud-trained models to identify hundreds of fish species and track coral health, and is being adapted for use in Fiji and Thailand. Recurrent neural network approaches to cyclone forecasting will be embedded in tourism evacuation systems. Sea-water air conditioning, in operation in the territory since 2006 and cutting cooling-related emissions by a factor of three or four, will become the default for luxury resorts. Most importantly, a Mā'ohi-controlled corpus and speech recognition system for Reo Tahiti, modeled on the community-governed Te Reo Māori work in Aotearoa-New Zealand, is plausible — but only if the Polynesian Assembly enacts a data-sovereignty regime in time. The negative pressures are equally real. English-dominant large language models will, by default, accelerate rather than retard the displacement of Polynesian languages in tourism interfaces. Generative tools already produce on-demand "authentic Polynesian" imagery, and Mā'ohi cultural institutions will almost certainly demand a sui generis intellectual-property regime to police it. Visitor-flow algorithms piloted in Venice and Amsterdam will be adapted for Fakarava, Rangiroa, Taputapuātea and Teahupo'o, raising the question of whether quota allocation migrates from communal deliberation into opaque optimisation pipelines.

The political envelope shaping all of this shifted decisively in 2023, when the pro-independence Tāvini Huiraʻatira party took thirty-eight of fifty-seven Assembly seats and Moetai Brotherson — a computer scientist by training — was elected President. A peaceful referendum on the territory's status is plausible within a ten-to-fifteen-year horizon. The most likely settlement by 2100 is neither full independence nor the current arrangement but something closer to a free-association model, with defense and currency remaining tethered to Paris while tourism marketing, education, fisheries, mineral rights and data policy sit firmly with the Assembly. The nuclear legacy — 193 tests conducted at Moruroa and Fangataufa between 1966 and 1996, including 41 atmospheric tests, with documented radiation exposure for essentially the entire population of the time — will, by century's end, most likely have evolved into a settled compensation regime, a publicly accessible archive, and a Mā'ohi-curated memorial tourism circuit through Hao, Tureia and Mangareva.

The shape of tourism itself, then, looks something like this. Annual visitor numbers settle in the range of 200,000 to 240,000, lower than today but generating comparable or modestly higher real receipts thanks to a deliberate high-value, lower-volume positioning. Bora Bora's share of overnights falls from roughly thirty percent today to perhaps fifteen or twenty, while the Marquesas, Australs and Gambier collectively rise from around five percent to a quarter of overnights. Cultural and heritage tourism — Taputapuātea, Marquesan tikis, marae circuits — and Tainui-Atea-branded marine experiences dominate the product mix. Reef-restoration and citizen-science tourism, descended from today's adopt-a-coral programs, grow into a genuine category. Pure beach-resort tourism declines. Most large resorts are cooled by sea-water systems, built on stilts, and protected by engineered shoreline defenses; a small number of Marquesan deep-water cruise piers exist; cruise itself stabilizes at around a fifth to a quarter of arrivals, dominated by small expedition vessels operating under zero-discharge-in-lagoon rules.

What this leaves is a resilient but smaller and re-indigenized tourism economy, anchored in the world's largest marine protected area, using AI as a working tool rather than a master, embedded in a politically autonomous but materially co-dependent relationship with France. The most important determinant of which version of this future actually arrives will not be modeling or technology but the choices being made now by the Polynesian Assembly, by community rāhui councils, by CRIOBE and Coral Gardeners, by the Académie Tahitienne and Tahiti Tourisme. The future of Te Ao Mā'ohi as a place to visit will, in the end, be defined less by external prediction than by what Mā'ohi communities themselves decide they want their tourism to be.