Despite technical efficiencies, the process of booking travel arrangements remains a very human task. Travelers have to manually find flights that work for their schedule, book hotel accommodations that work for their plans and price range, and estimate times of arrival while accounting for weather delays or possible layovers. Digital technologies have made this information more available and contextually relevant than ever before, but it still requires a human hand to comb through the data and make connections. However, as computers become more sophisticated and gain access to more data about travelers’ requirements—their calendar appointments, their membership programs, their travel preferences—these systems will soon be able to make these inferences for themselves.

“Can a system compute the best course of action to reach a destination based on a number of preferences?” asked Punchkick iOS engineer John Norton, “For example, you like leg room, you’re cost-conscious, you hate layovers. How can we design systems that take all this disparate information and arrive at the best solution?” This process typically necessitates human intervention, but it’s conceivable that an advanced search algorithm might process these variables with different filters or switches. But there are some kinds of travel plans that seem like they will always need a human eye.

“There are a few classic examples of where these problems can get really complex,” said Punchkick strategist Dan Cortes. “Imagine you’re trying to plan a multi-city trip through Europe. Your trip is going to take a few weeks, and in that time you want to visit a handful of the major capitals—London, Paris, Rome, and so on. You also have a few preferences—you don’t want to waste time on layovers, you prefer window seats, you’ll only fly certain airlines. What’s the best route for you to take through Europe, so that you see all the sights you want, all while keeping costs down and accounting for your preferences? Computers today would choke on that many variables.” To date, the only computer that can successfully balance this number of objective and subjective variables is a human brain. But computer scientists are hard at work building a processor that can match it.

The only computer that can successfully balance this many variables is the human brain. For now.

One of the most exciting fields in computer science is quantum computation, which might offer one solution to problems of this complexity. Quantum computers differ from classical computers in that, instead of processing instructions encoded in ones and zeroes—also known as “bits”—they accept instructions in the form of quantum bits—or “qubits”—which can be a one, a zero, or any combination of both at the same time. Building on the science of quantum mechanics, which governs how the strangest subatomic particles behave, qubits are an extension of a concept called quantum superposition, which explains how two particles can occupy the same space at the same time.

For computer science, qubits expand the possible range of information and variability for computers exponentially. Why does that matter? Quantum computers would not be faster across the board than classical computers we use today—they’re only better at solving very specific calculations, called optimization problems, many of which classical computers cannot even begin to tackle. “If you’re just doing simple things like browsing the web or playing music or whatever, a quantum computer would probably feel kind of slow,” said Cortes. “But for complex problems with tons of moving parts, they might be the only kind of computer that can solve them.”

So, that’s a lot of theory. What kind of experience would a quantum computer deliver for the end user? In the future, more sophisticated quantum computers could drive some of the web services and algorithms behind search products we use every day. The European trip scenario above is a common example of an optimization problem that quantum computers might be well suited for. “Someday soon, you’ll call up Siri and say something like, ‘I need to be in San Francisco by noon on Thursday,’ and she’ll just take care of it,” said John Norton. “Your calendar, your preferences, your airline rewards program, all taken into account on the backend. And Siri on your phone just says, ‘Okay,’ and books everything.”

“Someday, you’ll tell Siri ‘I need to be in San Francisco,’ and boom, she’ll just take care of it.” —John Norton, iOS developer

There are products already that aggregate data and simulate similar experiences, like Google Now for Android. Google Now accesses information like your Gmail messages, calendar appointments, or weather and traffic conditions to provide contextual and timely information right from the launcher. As services get more access to information, and as the backend computers processing the data become exponentially more powerful, this experience won’t need to be simulated—it will be real.

“The next step is for the system to take care of things without you having to ask,” said Norton. “If you schedule a meeting with someone in San Francisco, and add it to your work calendar, you could authorize your phone to make flight and hotel accommodations automatically, and charge your company credit card.” The idea takes the notion of a virtual assistant to new heights. “Someday, I expect that I’ll schedule something out of town, and the first time I’ll have to think about it is when the Uber arrives at my house to take me to the airport. My phone will already have my boarding pass, my hotel key, and everything else ready to go.”

Technology is becoming increasingly aware of its context, and increasingly proactive in making real-time data relevant to the user. As the supportive backend systems become more sophisticated, and as smartphone apps become better attuned to their users’ needs, these predictive experiences will leave the realm science fiction and enter the realm of Apple TV commercials.