The term AI or artificial intelligence has become ubiquitous in any innovation related conversation.
While the term AI is vast in scope and can include anything and everything which a machine can do in a human-like fashion, it is important to remember what AI is and what it is not.
So, first things first, let’s make an important distinction. The ability of a machine to perform a human task is not the same as a machine’s ability to perform a task in a human-like manner. This is important in understanding AI because in order for a machine to perform a task in a human-like manner, it has to have some cognitive ability, and this is where ‘intelligence’ enters the frame. A simple example of an AI application is a thermostat. Nowadays you don’t even need to explicitly programme it. The machine will learn from the data, environment and inhabitants prior choices and re-adjust itself.
The ability of a machine to perform a human task is not the same as a machine’s ability to perform a task in a human-like manner
Over the last decade or so, there has been enormous work and development in the field of AI, thanks to graphical processing units (GPUs) which can process enormous quantities of data in very little time, which was a huge limitation in the past. Two key concepts emerged from this development; machine learning (ML) and deep learning.
Simply put, ML is the ability of a machine to learn from its previous experience and improve its performance with every successive iteration, much like humans do. A good example of ML are the route navigation apps on smart phones. The apps are able to predict with a high degree of accuracy how much time it will take a person to get from point A to point B factoring in real-time information about traffic conditions, speed limits, alternative routes and a host of other inputs. As these inputs change, the predicted arrival times self-adjust.
Deep learning, on the other hand, is a highly evolved field of study in machine learning where a machine mimics the human brain in processing information using something called neural networks. Just as a neuron in the human brain is connected to multiple other neurons creating multiple pathways for information exchange, deep learning uses more or less the same method to probe and seek data across multiple levels of data.
4 applications for science in travel
The evolution of this science has numerous applications in travel.
Image recognition: It is conceivable that identity validation at airports and hotels for check-in /out will be largely done by machines in the not so distant future. In fact, Delta airlines has already begun a Pilot at Atlanta Airport wherein a traveler’s security screening and identity validation is done via cameras. Once this technology becomes more affordable and main stream, it poses a question; Do we still need a front desk at hotels? Check-ins for most part are already happening via mobile devices.
Virtual / Augmented Reality: We can now feel firsthand what is it like to be in a room or conference space or even bungee jumping. We will no longer be limited to making buying decisions on someone's recommendations or peer ratings but be able to make a choice based on highly personal preferences.
Chatbots: Although chatbots seems like yesterday's news, the Natural Language Processing (NLP) technology which powers the dialogue is becoming increasingly sophisticated and capable of doing far more than simply replying to FAQs. Chatbots are now able to conduct a complete conversation autonomously with users and make transactions. We know a lot of travel is still booked by human agents and so is customer support. These sophisticated chatbots significantly reduce the costs associated with both these activities.
IOT & Personalised content: Some hotels are actually testing a cloud-based technology whereby guests can wear an RFID enabled wearable device, which allows them to walk past any vending outlet at the resort, and buy any product or service without having to stop to make payment. Smart sensors automatically identify travelers via the wearables and associate the purchase with that account. In future this technology can also be leveraged to personalise your experience while you stay at a hotel or travel through an airport.
There are plenty of other exciting things which are possible today using technology which may have seemed like sci-fi stuff a decade ago. However there is one thing which, for the moment at least, machines can't take away from humans, and that is empathy and care. That is where the travel industry needs to focus its energy, and training the workforce of tomorrow would be a good start.
Back to news articles