Canberra Times
Using artificial intelligence (AI) to plan a holiday is all fun and games until you’re standing in a remote border hut in Zimbabwe, explaining to an immigration officer that a chatbot swore your visa was valid.
Australians are increasingly using AI tools to research and plan overseas trips, but a misplaced reliance on hallucinated information is leaving holidaymakers paying for costly mistakes.
Aussie traveller Max Phelps learnt this lesson the hard way, after planning a multi-nation African safari for his family of three.
“We landed in Victoria Falls in Zimbabwe and we had the KAZA visa, which we believed was the right visa to get for our trip,” Mr Phelps said.
But when an immigration officer asked about their plans, he informed the family that they had the wrong visa and would need to reapply when they attempted to re-enter the country days later.
Mr Phelps balked at the immigration officer, assuming that he was trying to sell them an unnecessary visa, but later realised that he had been correct.
A popular AI chatbot helped Mr Phelps design and optimise his itinerary, finding locally-owned safari tours, determining the best flights and planning a trip route.
But when he asked the chatbot which visas he would need to navigate Zimbabwe, Zambia and Botswana, the AI responded confidently that a KAZA visa would be ideal.
It even reassured Mr Phelps that the visa was valid for unlimited entry across 30 days, but at the border hut in Zimbabwe, he learnt that AI was wrong.
“It’s expired because you’ve been out of the country for two nights,” the border officer told him.
Luckily, the issue was easily resolved. After half an hour and $200 for new visas, the family was on their way again.
But it’s not just the Phelps’ family. Generative AI platforms are also misleading travellers about baggage allowances and connecting flights, resulting in chaos at the airport.
Adam Ewart, the chief executive of luggage delivery service, Send My Bag, said research by the company found that almost 40 per cent of Australian travellers were using AI to plan trips.
“AI is very good at turning complex information into a clear, simple answer,” Mr Ewart said. “That convenience is also the danger.”
“It can make advice sound definitive without showing where the information came from, when it was last updated or whether an exception applies to that particular traveller.
“People may also assume AI is checking official government sources in real time when that is not necessarily the case.”
Older travellers were slower to adopt the technology, while more than 50 per cent of Gen Z and Millennials were starting to use AI for travel planning, Mr Ewart said.
With younger travellers leading the charge, Mr Ewart warned familiarity with a technology should not be mistaken for certainty that its answers are correct.
“Being highly confident using AI is not the same as being able to identify when it has made a mistake,” he said.
“Use AI like a well travelled mate: ask it for ideas and use it to prepare, but don’t treat it like an airline, embassy or border official.
“AI can do the homework, but it should never be allowed to mark its own exam.”
