Artificial Intelligence is no longer just hype: it is a reality. Artificial Intelligence-based approaches, such as Natural Language Processing (NLP), Machine Learning (ML) and Deep Learning, are becoming more realistic within the technology industry. We have today very efficient NLP engines as well as powerful ML and deep learning algorithms available. In a recent article on WIRED, I remember reading about the death of code (programs and programming) and how we will soon be training systems like we train our pets.
Machine Learning involves learning from examples and experiences: it’s all about digesting huge volumes of data. IBM and Memorial Sloan Kettering are training Watson in Oncology using the massive amount of patient medical records across the world. Watson learns from how doctors treat cancer patients worldwide, similar to how a medical student learns but on a larger scale.
Another example of machine learning can be found in Japan: here, farmers cultivate fresh and crispy cucumbers, with many prickles on them. Straight and thick cucumbers with a vivid color and a lots of prickles are considered to be of premium grade. Each cucumber has a different color, shape, quality, and freshness. Cucumbers are sorted into nine different classes based on size, thickness, color, texture, small scratches, whether or not it is crooked, and the number of prickles on it. There is no well-defined instruction set for classification of cucumbers.
Makoto Koike studied this problem while helping his parents to classify the cucumber on their farm. Using Google’s TesorFlow-based machine learning algorithm, Makoto developed a system that learns from the precise way his parents sort their cucumbers. To do this, he trained the system by using 7,000 images of cucumbers sorted by his mother, and now the system classifies cucumbers with a better success rate at rapid speed.
At Capgemini, we have been using IBM’s Watson to help us improve effectiveness and efficiency in the resource supply chain.
The AI wave will take the industry by storm, impacting almost every business and transforming the current technology climate. In addition to ML, we need to quickly transform our businesses using Al-based approaches supported by NLP, Optical Character Recognition (OCR), speech recognition and image recognition. There are three key trends in favor of the current technology service providers and their employees:
- Global spend on technology is increasing. Therefore, technology companies will increase their size and market share by adapting to these new ways of working.
- The current availability of Al technology worldwide is less than the amount the world needs, so the companies and individuals who quickly expand their Al capabilities will shine in the market. For those interested in Al, this is the perfect time to advance their skills to become market leaders.
- The transformation in the industry has resulted in the marginalization of the CIO role in business and the purchase of technology services by business buyers. This gives an edge to business-oriented business teams.
However, reacting to this new technology demand for AI brings a few challenges: First, change can only happen when company stakeholders believe in it. Unfortunately, many employees and managers do not believe in the power of AI until they experience it themselves. It is those who believe, develop the required skills, and embrace this evolution that will flourish. Secondly, companies must address how to deal with the possible cannibalization of existing revenue in order to adopt these new technologies. Finally, the scarcity of skill in the technology world makes it ever more difficult to build and expand AI capabilities. Due to an industry boom over the past 20 years, a sizeable percentage of existing staff have obsolete skills and cannot be taught the necessary new ones.
This is going to be an interesting future journey for the industry.