Airlines are using AI to set ticket prices. Here’s how travellers can protect themselves | Canada Voices

If you’ve ever ticked “agree” on a terms-of-service notice without reading it, mindlessly clicked “accept” when asked if you’d like to enable browser cookies or simply logged into your loyalty program, those decisions may soon mean paying hundreds of dollars more for flight tickets.

Delta Air Lines Inc. has drawn the ire of some U.S. lawmakers over concerns that airfares could increase after airline president Glen Hauenstein said in an earnings call that about 3 per cent of Delta’s domestic ticket prices are set using artificial intelligence.

The technology from Israeli startup Fetcherr offers airlines the ability to use factors such as customer lifetime value, past purchase behaviours “and the real-time context of each booking inquiry” to create “a truly personalized offer.”

“We like what we see, we like it a lot,” said Mr. Hauenstein of revenue impact, adding that he hoped to expand the pricing system to 20 per cent of tickets by end of 2025.

While Canadian consumers have slightly more protections than Americans regarding the personal data that companies can gather and how it’s used, regulations and enforcement are often a few steps behind when it comes to AI and its rapid advancements.

In Canada, carriers such as Porter Airlines acknowledge that they rely on dynamic pricing and that AI is part of some business processes. But it isn’t used for personalized pricing for passengers.

Whether that ever becomes the case, there are still ways consumers can protect themselves.

In Canada, the Personal Information Protection and Electronic Documents Act, a federal privacy law, requires that businesses get “meaningful consent” before collecting, using or sharing consumers’ personal information, said Terry Cutler, a Canadian information security strategist.

“In theory, that should give consumers a bit more protection than what’s currently in place in the U.S., where privacy laws are fragmented and often state-specific.”

But the act is outdated and doesn’t clearly cover how AI is being used to create dynamic or personalized pricing models.

“So, while we’ve got some protections, there’s still a lot of grey area, especially when your data is being used in ways you didn’t expect, like setting prices based on your browsing habits or device type,” said Mr. Cutler.

Implied consent is also often sufficient for some personalization, such as being shown ads for Ottawa Senators tickets after Googling the team, according to Kris Klein, founder of nNovation LLP, which specializes in privacy and information security issues, and part-time law professor at the University of Ottawa.

Express consent is required when the profiling uses sensitive data or “goes beyond reasonable expectations” – for example, if an advertiser targets someone after they search for a medical condition, such as sleep apnea.

But AI-driven personalized pricing, the sort adopted by Delta, may already be in use to some extent in Canada as enforcement of the law relies on complaints and “people don’t always really know what’s being done to them,” said Mr. Klein.

While true personalized pricing is still rare in Canada, components of it are emerging, said Ceren Kolsarici, director of the Scotiabank Centre for Analytics and AI at Queen’s University’s Smith School of Business.

Air Canada uses dynamic pricing for its fares and Aeroplan redemptions, adjusting in real time based on demand and customer segment, for example. While it’s not exactly personalized, it’s algorithmically optimized and increasingly granular, she said.

In June, the software company PROS Holdings Inc. shared a customer testimonial and video discussion with Porter’s director of revenue management systems, referencing how the airline was using a feature called Willingness-to-Pay (WTP) and Porter’s aspirations in adopting the real-time dynamic pricing tool add-on.

According to PROS, these tools so far only rely on historical data – such as how full a flight was last year on the same route and date – as well as contextual factors such as major events.

But the tool can, for example, anticipate that business travellers may book closer to the departure date than tourists and are willing to pay more, allowing airlines to adjust prices accordingly.

To limit how much data is being used against you in pricing, now or in the future, Ali Dehghantanha, a Canada Research Chair in Cybersecurity and Threat Intelligence at University of Guelph, recommends not logging into airlines’ loyalty programs until the moment of purchase.

“When you browse while logged in to your frequent flyer account, the airline’s pricing engine can instantly tie the search to your full profile – travel history, average spend – giving it extra signals about how much you’re likely to pay,” he said.

He also recommends using virtual private networks to hide your location and relying on third-party fare trackers such as Hopper that query prices through server-to-server communication.

Watchdog says Ottawa should allow foreign-owned airlines to fly domestic routes to boost competition

Using incognito or private browsing when shopping or booking travel, which helps avoid price manipulation based on your history, is a must, but it’s also important to consider the browsing history of other people in your household.

If one spouse searches for plane tickets at the same internet protocol address, it can impact the price the other spouse gets as well.

He also recommends clearing your cookies and browser cache often to wipe stored data companies use to track you across sites.

Don’t forget to consistently audit privacy settings for your apps and browsers. “You’d be surprised how much you’ve unknowingly allowed,” he said.