Artificial intelligence (AI) may conjure up images of intelligent systems rising up to enslave us, with sadistic driverless cars taking passengers on terrifying joyrides and malevolent smart fridges refusing to order milk.
But in reality, AI is already here; many of us just do not realise it yet. Through the development of machine learning, the technology industry has created smart systems designed to help, not hinder, people.
Rather than send murderous robots back in time to destroy us, machine learning sits behind the scenes in the form of algorithms that learn from data to better deliver services offered by software applications.
This is the direction that Forrester principal analyst Diego Lo Giudice believes AI is heading: "While some AI research still tries to simulate our brain or certain regions of it - and is frankly unlikely to deliver concrete results anytime soon - most of it now leverages a less human, but more effective, approach revolving around machine learning and smart integration with other AI capabilities," he said in a blog post.
One example of this is Office Graph, created by Microsoft. Office Graph sits behind Office 356 and maps the relationships between people, content and activity across the company's productivity software suite.
Office Graph uses these relationships to serve up documents, files, email messages, and other elements in Office 365 that are relevant to the work the user is carrying out, with the goal of making the user more productive.
Another example is Microsoft's Skype Translator, which uses machine learning capabilities to improve the accuracy of its translation of spoken languages, something CEO Satya Nadella likens to "magic".
"The one fascinating feature of this is something called transfer learning. What happens is you teach it English and it learns English. Then you teach it Mandarin, and it learns Mandarin, but it becomes better at English," he said.
"Then you teach it Spanish, and it gets good at Spanish but gets great at both Mandarin and English. Quite frankly none of us knows exactly why. It's magic."
Not wanting to be left behind, Google recently announced an update to Google Translate to provide real-time translations that use machine learning to deliver better results the more it is used.
Big data analytics
But, while machine learning might improve the performance of consumer-level services, it is in the enterprise world that it really shines.
Big data is a big business, but trawling through hundreds of gigabytes of often unstructured information can be a lengthy and resource-sapping process. Add machine learning into the mix and that process becomes a lot easier.
Rather than rely on data analysts to come up with patterns and predictions based on samples taken from vast reams of diverse data, machine learning can crunch an entire set of big data and use algorithms to come up with accurate and tailored results.
Given the ability of computer programs to crunch data faster than the human mind, machine learning offers data-driven organisations a much faster and efficient way to exploit big data.
Furthermore, by having algorithms take care of the resource-intensive and complex work, machine learning means big data analytics becomes accessible to business users, rather than just highly trained data scientists.
New tech battleground
Creating machine learning algorithms and systems is a complex and costly task, so most enterprises will prefer to buy in external machine learning-based services to use with their harvested data.
Some of the world's biggest tech companies are now looking to tap into this burgeoning market. IBM, for example, has built an entire machine learning ecosystem around its Watson supercomputer.
This ecosystem includes the Watson Discovery Advisor, which seeks out hidden connections in big data, and Watson Analytics, a cloud-powered service that adds natural language recognition to data analysis.
Microsoft also has a stake in cloud-based machine learning with its Azure Machine Learning service, which gives users the ability to deploy machine learning onto their data sets from Microsoft's cloud platform.
Being based in the cloud, Azure Machine Learning allows users to select custom analytics algorithms that best suit their business from a web browser, thereby bypassing the need to install equipment or resource-intensive applications.
Meanwhile, Amazon has entered the fray with its cloud-powered Amazon Machine Learning service, which aims to give developers access to machine learning without requiring them to learn complex algorithms.
Through the use of application programme interfaces (APIs), Amazon's service provides developers with a way to plug machine learning into their apps, bypassing the need for them to hard-code algorithms into their software.
"You can build and fine-tune predictive models using large amounts of data, and then use Amazon Machine Learning to make predictions (in batch mode or in real-time) at scale," said Jeff Barr, chief evangelist at AWS, nothing how this will widen the ability for business of all shapes and sizes to use machine learning.
"You can benefit from machine learning even if you don't have an advanced degree in statistics or the desire to set up, run and maintain your own processing and storage infrastructure."
From insight to action
The majority of machine learning services use predictive analytics to serve up insights and forecasts from big data to business decision-makers.
But, in future algorithms could tell users what actions to take, rather than just make predictions, effectively moving from predictive to prescriptive analytics.
This is the direction in which cloud-based machine learning firm Inside Sales is heading, with the help of backing from Microsoft and Salesforce. It uses its proprietary algorithms to sift through a vast collection of aggregated data harvested from its users to provide salespeople with guidance on how best to pursue a lead or close sales with potential customers.
Inside Sales' machine learning technology works by harvesting data from each of its users' customer relationship management systems and databases, anonymising that information and analysing it against similar data gathered from other salespeople and their databases.
That machine learning is then applied against a salesperson's custom parameters to recommend actions that will improve their chances of securing a sale.
For example, through machine learning data analysis, Inside Sales can discover that people working in finance within London are more likely to answer sales calls at 3pm rather than 11am.
Furthermore, just as Skype Translator learns as the data it is fed increases, as more firms sign up to services like Inside Sales, the larger the database it has to anaylse, in turn helping improve its analytical capabiltiies and insights.
The rise of the truly intelligent machines may be some way off, but there's no doubt smart machines and systems are going to play an ever increasing role in the way enterprises function.
Q3 losses reverse Q2 gains
FBI briefing US companies to dump Kaspersky, claiming intelligence prove it a 'threat to national security'
Kaspersky rejects FBI accusations that its products are a 'threat to national security'
But breached contractor says that it simply didn't have that much data
EE follows Three in threatening legal action against Ofcom - but for entirely different reasons