How Soon Will AI Revolutionize Our Industry

By Kim Macaulay, IATA Senior Vice President, Information and Data

I'm often asked about the extent to which AI is revolutionizing aviation. The easy answer is "a lot", but the more complicated answer is that the extent of AI penetration into airline operations differs dramatically depending on which part of the business we are talking about.

Let's consider safety, the number one priority of the aviation industry. Scaling AI for aviation safety is less about technical maturity, and more about evolving organizations to absorb AI as it becomes more capable to analyze vast amounts of data for prediction, text classification, efficient image and speech recognition. Here, AI provides the advantage of processing a large amount of information to support faster and more accurate decision-making in the identification and management of safety risks.

New Use Cases

There are many other areas where AI is already being implemented by airlines. Carriers are investing in training their own AI systems to help calculate things like route optimization, predictive maintenance schedules, sales efficiency, revenue maximization, and forecasting. Also, AI-powered Chatbots offer opportunities to improve customer service. Of course, not everyone wants to talk to a robot, but a chatbot is probably an upgrade on searching through a website for standard information, and if it can handle basic queries, that frees up human customer service agents to assist with more complex issues. On balance, I believe chatbots will lead to a better customer experience, not simply be a cost-cutting exercise.

AI Agents

Perhaps the most interesting current application of AI is in the realm of 'Agentic AI' - where AI 'Agents' act autonomously on our behalf to fulfil ever-more complex tasks. For example, each year, IATA publishes the Dangerous Goods Regulations - a massive document covering thousands of rules for the safe handling of goods. With Agentic AI, a freight forwarder Agent could chat to a shipper Agent and a cargo carrier Agent to arrange a shipment, and ensure it is correctly labelled and packed in accordance with the DGR rules, with only the minimum of required human oversight.

To those who might question if there is a safety risk, remember that as AI improves, it will progressively reduce mistakes, whereas most humans in the chain have probably already "peaked" their performance. Human oversight and accountability are vital, but humans are not infallible. In fact, machines will ultimately make fewer errors.

How fast will we see these AI applications implemented? It's happening right now. Airlines are making heavy investments in data and Large Language Models (LLMs). The first step is to ensure that trusted frameworks for sharing data are developed. It's obviously vital that commercially sensitive information can't be hacked, and that important operational data can be shared in confidence between industry partners.

Reducing the Cost

Cost is still a barrier. Airline margins are much thinner than many other large business sectors, which reduces the potential investment in expensive cutting-edge AI. However, as a 'hyper-Moore's law' helps dramatically increase computing power and costs reduce, more airlines will be able to access AI applications. Presently, some 20-40% of coding for airlines is being done by AI, but this area of development is likely to ramp up, with subsequent huge productivity gains. Airlines can see the potential opportunities of this - for example, IAG and Emirates have announced partnerships with OpenAI.

Ultimately, we can expect to see shared AI models working between airlines, airports, ANSPs and ground handlers to optimize the efficiency of the entire aviation chain. Baggage and cargo, air traffic management, slot allocation, and passenger biometrics are just some of the obvious applications that will further enhance operations and the passenger journey experience.

Bad news for employment?

Perhaps the biggest question facing businesses and employees on the cusp of this AI revolution is "Will this lead to massive job cuts?" My honest belief is that, at least in aviation, this will not be the case. There will be some traditional 'back office' roles and support functions that AI will take over. But new jobs are likely to be created as well. And with passenger numbers set to double by 2050, we are still going to need more pilots, more cabin crew and ground staff. We all still want and expect that human touch - the smile of the purser at the gangway, the reassuring voice of the pilot updating us at 30,000 feet. That's something no AI will be replacing soon.