A featured contribution from Leadership Perspectives: a curated forum reserved for leaders nominated by our subscribers and vetted by the Construction Tech Review Advisory Board.
Product & Technology
Starting your AI Journey
Agustinus Nalwan
Artificial Intelligence (AI) and Robotics Process Automation (RPA) are no longer just buzzwords, they are hot discussion topics at conferences, many research papers have been written about them, and third-party vendors have already built tools and services to help companies acquire these capabilities.
There are fundamental reasons why RPA should be implemented by an organisation. AI, which is responsible for the success of RPA, enables an organisation to perform tasks faster with less human effort, at lower costs and with a better quality outcomes – most of the time.
For example, an insurance company could use a chat-bot and a recommendation system powered by AI to personalise an insurance quote 24/7, which may help to reduce operational costs.
AI technology also enables smaller organisations to deliver the same bandwidth. In fact, the bigger you are, the more alert of smaller players you should be.
At carsales, even though we are the largest online automotive, motorcycle and marine classifieds business in Australia, we are always actively monitoring our competitors, big and small.
AI technology also enables smaller organisations to deliver the same bandwidth
When I speak at conferences, one of the most popular questions that I get asked about AI is - how do you acquire AI capabilities?
Firstly, I recommend an organisation really considers how they’ll use AI and their need for a data scientist.
If your core business is building autonomous drones or the next AI frameworks or tools, then yes, by all means you need a top data scientist. In contrary, if you are just wanting to forecast flight ticket costs for a flight booking system, all you need is a software engineer who knows how to use an AI/RPA tools or cloud services.
Understanding the data science technique behind the tool you are using is a nice to have, but it is not critical. Cloud services are rapidly rolling out their AI tools, and they are getting easier to use, therefore data science experience is not required to operate them.
In addition to assembling your AI team and commencing the build your RPA, another important step you should take is the integration of the technology you are building with your existing workforce.
At carsales our strategy is to plan the integration process even before we start building the AI. It is inevitable that AI will take over tasks normally performed by staff members, but we train our people to direct and train the AI system.
It is also important to communicate changes early to staff to ensure they feel comfortable and secure with the integration of the new AI processes.
The current AI being built is very good at performing a task in one knowledge domain, it’s deep but very narrow. For example, current AI fails to identify cars that it has not seen before, even if presented with a Toyota Corolla and the word “Corolla” is clearly visible on the car, which should be sufficient for the AI to guess what the car is with a high degree of accuracy.
Another limitation of current AI is the lack of knowledge abstraction. This makes it hard to train AI without lots of labelled training data, which is a problem currently experienced by many.
However, there are many benefits of using AI and I believe boosting productivity has to be one of the most significant.
