What Does Machine Learning Mean for the Enterprise?
Machine learning algorithms are already being used in a wide variety of applications – from smartphones to deep space. Indeed, NASA recently discovered a planet orbiting a star (much like our own sun) over 2,500 light years away by using Google’s machine learning technology.
The rapid rise of the Internet of Things (IoT) and big data applications are only serving to accelerate this new era of automation.
While these are exciting times, it’s important not to get carried away with all of the hype. AI is not going to result in mass job losses or, as some tech luminaries believe, kill us all. In fact, it is predicted that by 2020 more jobs will be created than lost as a result of AI and machine learning.
It’s early days for machine learning in the enterprise, but what is clear is that this new field of science offers organisations a host of opportunities and we are only just scratching the surface with regards to what’s possible.
It is predicted that AI augmentation – that is the combination of human and artificial intelligence – will result in $2.9 trillion in business value and recover 6.2 billion working hours by 2021.
What is machine learning, really?
While the algorithms behind machine learning are complex, the concept is relatively straightforward. Machine learning examines large amounts of data looking for patterns, then generates additional code that lets it recognise those patterns in new data. Your applications can use this generated code to make better predictions. The more data the algorithm has access to, the smarter it becomes.
According to Gartner, machine learning is currently best suited to highly repeatable tasks where large quantities of data can be analysed for patterns. Fraud detection is an excellent example.
Like any new technology, machine learning is meeting cultural resistance in the enterprise. Many organisations have processes that are entrenched in the company and there is a lack of willingness to hand over control to something people don’t fully understand. Many tools are also in their infancy which inevitably leads to a degree of risk; and there is also a shortage of skills, particularly in the area of data science.
According to the Confederation of British Industry (CBI), half of businesses think AI will soon ‘fundamentally’ transform their sector but only a third believe they have the skills and capabilities necessary for adoption.
Cost is also a factor, with typical early adopters being global players or digital natives like Facebook, Google and Amazon.
With so much investment pouring into the field, it should come as no surprise that the technology is becoming more accessible all the time. In many cases, SaaS providers are now implementing machine learning capabilities into their solutions. For example, Box is adding image, audio and video machine learning capabilities to its collaboration and storage platform, so that customers can create their own AI processes. Office 365 draws on machine learning for several of its productivity and collaboration solutions such as Delve.
Vendors are also developing tools that make the creation of machine learning applications much easier. Azure Machine Learning Studio is a browser-based drag-and-drop authoring environment, allowing developers to easily and quickly create machine learning applications. The arrival of such tools will undoubtedly improve accessibility and help early adopters explore the capabilities and limitations of machine learning.
Fourth Industrial Revolution
Machine learning may still seem relatively inaccessible to many, but there is no question that this technology is going to underpin future digital transformation efforts and play a pivotal role in the much-heralded ‘Fourth Industrial Revolution’. Just because NASA is using it to find new planets, it doesn’t mean you need to be a space agency to start making the most of this transformational technology today.
Learn more by reading this eBook: The 5 Keys to True Digital Disruption