How Is Artificial Intelligence Developing In The Enterprise?

IBA Group
Mark Hillary

Artificial Intelligence (AI) has moved from science fiction into the enterprise in recent years. Many companies are using AI systems today to intelligently analyse large volumes of changing data and to notice or predict patterns. Typical business uses today include examples such as:

  • Rail operators predicting train delays before they happen because the AI system can extrapolate from small delays to predict the impact on the entire network.
  • Customer service agents being advised on how to help customers by systems that know the answer to every question a customer has asked in the past.
  • Alexa knows how to answer your question because it immediately processes your voice and determines what you are asking before creating an answer.
  • Netflix knows which movie you might want to watch because they know your past behaviour and how similar customers have also behaved.

AI really is all around us today, in the enterprise and as consumers of services. In the present environment it would now be unusual for any company to not be exploring how AI can improve their business.

But AI does have one a fundamental flaw, it is always limited to working on a very specific problem. This means that you can have a very complex system that knows everything that your customer may ask when they call for help or the system may understand how to play chess or Go, but these individual tasks are all that it can do. There is no inherent awareness of the environment around the system – although we use the term intelligence, it’s not really aware or sentient. An AI system that can play Go cannot plan the best route on a map.

This means that the system can only solve the problems it was designed for. Some might argue that this is a benefit, because it means that however good our AI systems get, they never move into the realm of awareness and all the problems that a conscious system might create.

A recent experiment by IBM has demonstrated that AI is developing rapidly though. They demonstrated how an AI system could be asked a random question and it would then have the ability to debate that subject. For example, in the video clip that I watched the system was asked if pre-school facilities should be subsidised by the government. It gave a response, arguing why subsidies are useful for 4 minutes.

This system has been pre-loaded with information on millions of subjects and objects. It’s stuffed full of encyclopedia content and research. But even with all this data it is quite an achievement to turn that into information and then a coherent argument.

Essentially this system is starting to show that perhaps an Artificial General Intelligence might be possible. It would need to be pre-loaded with an enormous amount of general data, and then would need a Machine Learning system to continue learning, but it is looking more feasible than even a year or two ago when Elon Musk started warning that we are heading for an ‘AI apocalypse’ because the machines will eventually have more intelligence than the humans.

I don’t think we will be seeing many business case studies featuring general intelligence just yet, but AI in the form we already know it will certainly be more important. AI is offering companies a chance to identify patterns and trends they could never see manually and this will be a strong source of competitive advantage in the next few years.

Artificial Intelligence (AI) has moved from science fiction into the enterprise in recent years

Business Application Support With Artificial Intelligence

IBA Group
Mark Hillary

I mentioned in my last blog that the very nature of jobs and employment is changing today because of emerging technologies such as Artificial Intelligence (AI). Tools such as AI in business application support, are fundamentally changing how corporate processes function – and this changes the skills that people in professional jobs need. This change is prevalent throughout many parts of the modern organisational structure, but the most important area of change is probably enterprise decision support.

The problem is, how do we get from here to there? Many organisations are overrun with data. They have so much that they just don’t know what to do with it all. Some have tried to focus on data analysis, creating a data-driven approach to their business, but even those who have moved in this direction need to rely on the skills of their data scientists to try turning data into information.

As data analytics evolved, it was clear that enterprises were changing from a backwards-facing approach where they analysed past events and described what happened, to an ability to be predictive – trends and patterns that help to predict future behaviour could be found. A similar change needs to take place in enterprises using AI today.

Today, the possibilities for AI-enabled decision-making are more prescriptive, with AI providing enterprises not just a look into the future, but also key diagnostics and suggestions on potential decision options and their payoffs. Such highly evolved applications of AI can help businesses make decisions that can potentially exploit more business opportunities, while averting potential threats much earlier.

This is the real opportunity for AI in the enterprise. To create the opportunity for automated decision-making, but where the system is designed to learn, unlearn, and relearn as often as required. Insights from automated AI business application support should be powerful because the system is trained to learn where insights may be found.

I believe we will be seeing many more AI business application support systems in future – its no longer a technology of the future. These are systems that are already installed and in use in enterprises globally – have you explored what AI can do for you and your decision-making processes?

When Will Your HR Team Start Learning To Code?

IBA Group
Mark Hillary

There are many emerging technologies that are not only changing the workplace, but changing the way that jobs are structured and the skills that modern (or future) employees need. I believe that three of the most important changes taking place at present include:

1.    Artificial Intelligence (AI) in the workplace
2.    Remote working
3.    Creating tribes

There are technologies and systems emerging, such as AI in business application support, that are fundamentally changing how corporate processes function – and this changes the skills that people in professional jobs need. Employees need to become comfortable with the idea of AI delivering performance feedback, personal development, coaching and evaluation. This offers many advantages to both employee and employer, but it can still face resistance by some employees, especially when they feel it will change their job.

Forbes magazine recently published data from a study by the Center for Effective Organizations at USC Marshall School of Business. The study suggested that only 37% of employees would share innovation or automation ideas if they believed they would have to do different work as a result of such technology being implemented. However, when employees believed the technology would help make their job better, 87% of them said they would share innovation ideas with their employer.

Both AI and employees will help companies to reengineer their processes, but with AI exploring how to optimise systems there is an opportunity to change processes without the natural reluctance of the employees.

Remote working is increasingly a reality in many industries. Customer service companies are now actively promoting the Work-At-Home-Agent model instead of increasingly large contact centres. Companies with a large number of home-based employees can dramatically reduce costs for office estate and more easily scale up and down as the business requires.

The need to create tribes is partly related to the trend for home-working, but it is also linked to our increasing use of social networks. As we see people less in real life and more in virtual spaces, such as social networks, it becomes more important to be more methodical about socialising – both in person and virtually.

All these changes in the way we work are related back to the increasing intelligence of systems that can help us to perform more effectively at work. We are now reaching a point where coding skills are becoming useful for employees in almost any professional job – accounting, HR, and law companies will all be using AI business application support and this means that professionals need to learn how to manage their virtual tools.

A great change is coming soon. It’s not that every job will vanish as many are automated, but those that remain will become more interesting and more technical – the HR team needs to start coding soon!

AI Is Bringing Brands And Consumers Closer Together

IBA Group
Mark Hillary

Artificial Intelligence (AI) is quite a pervasive technology in the present-day environment. Even regular consumers with no technical knowledge are becoming aware of AI and are comfortable interacting with these systems. Examples are all around, from Siri on the Apple iPhone to the movie recommendations made by Netflix and song playlists on Spotify.

But there are many other ways in which consumers are beginning to interact with AI systems and many of them are not so obvious, at least to the end consumer of the services. Think of self-driving cars. They may not be common yet, but they are being tested all over the world and they rely on AI to constantly monitor the environment outside the car and to decide what to do next to keep the vehicle safe.

AI can also help to predict what people will do in the future. Facebook can tell if you are likely to take your own life based on recent posts. Stanford University trained a system to detect if you are gay or straight based just facial photographs. The HR system designed by IBM can predict who is likely to quit their job. The implications for these insights are fairly clear – imagine what an insurance company or government could do with this data.

Perhaps more positively, there are now investment algorithms that outperform regular investment managers and AI-powered disease diagnosis means that your virtual doctor will be aware of any relevant research and drug trials – even if it was just published yesterday.

Most consumers will be largely unaware of these developments, but there is one area where people are creating a huge demand for greater investment and research into AI systems and that is personalisation – the interaction between consumers and brands.

Years ago it was Amazon that really started this wave of personalisation by offering deals or recommendations based on the specific shopping behaviour of the individual customer. This was extremely innovative at the time because most brands could only ever offer the same deal to all customers at the same time. Now this is commonplace and expected. A clothes retailer needs to know what the customer likes, dislikes, their shopping history, and what they have browsed and lingered over in the past. All these insights would be impossible for a person, but an AI system can figure out what to offer the customer – either as a recommendation or as a special offer – and ensure that the offer is made at exactly the time that the customer is most likely to respond positively.

Now these personalised insights are not only becoming more common, but customers know that brands have the data so they are expecting greater personalisation. Customer demand is creating a wave of IT research and development. AI is moving quickly from being interesting and innovative to becoming essential for brands across many industries and it is customer expectation that is driving this change.

Digital Transformation Is Redefining Business Models

IBA Group
Mark Hillary

The technology industry is in an interesting place right now. Industries across the world are facing a wave of digital transformation that is often redefining their entire business model and value proposition. In addition, the increased strategic focus on customer experience (CX) as a boardroom priority is creating a wave of investment in emerging technologies. There has never been a better time to be closely connected to the IT industry because IT is redefining so many industries at a scale that nobody has ever seen before.

Customer expectations have changed dramatically in recent years meaning that companies need to deploy every tool possible to increase both sales and loyalty. The Business to Business (B2B) environment research by Salesforce shows that 80% of customers are influenced by the way that a company is able to understand their individual needs. There is a clear need for organisations to explore how digital transformation and the use of more innovative technologies can create a better experience for their customers.

There are many examples of the changing digital environment and how emerging technologies are changing the interface between companies and their customers, for example:

1. Bots; intelligent chat bots are able to handle simple customer questions automatically deflecting calls that would need to be answered in a contact centre.
2. Automation; the use of Robotic Process Automation (RPA) to automate manual processes makes life easier for employees and increases productivity.
3. Data Analytics; the ability to personalise service for customers because the system knows exactly what they like, when they like to buy, and which channels they prefer. Companies can use data insights to show their customers that they understand exactly what the customer wants.
4. Immersion; Both augmented and virtual reality systems are being used for solutions such as allowing customers to experience a luxury hotel before booking and how to find products inside a store.
5. Self Service; customers are increasingly searching for help online before ever asking a company for help with their products so there is an increasing need to create intelligent content that answers customer questions in locations such as Google.

It’s clear that the combination of increasing customer expectations and the emergence of these new technologies is driving a wave of digital transformation. More than ever, companies are turning to their technology partners for answers. Technology is no longer just a service that supports the business, technology is rapidly transforming how organisations function and redefining business models. Technology is rapidly becoming the most important driver shaping how organisations function and define a business model today.

The Six Key Factors For Success In Digital Transformation

IBA Group
Mark Hillary

Digital transformation is a business strategy that has been increasing in importance in the past few years. The ubiquitous use of smartphones, access to fast mobile Internet, and the app store concept have all combined to create a platform where established companies can offer online services and new companies can go to market with innovative ideas extremely quickly.

In some industries, such as financial services, there is an arms race taking place. New companies are launching services that are free of the technical legacy a large company, such as an international bank, needs to manage. Freed from many of the traditional requirements – such as a chain of retail branches – these new services can offer better prices to customers and be entirely designed around the needs of the customer, not the legacy systems of an established company.

It’s clear that companies across all industries need to be exploring how digital tools can improve their service to customers, but digital transformation projects can be risky. If a company bets on the wrong type of service then they can quickly become irrelevant in their own marketplace. Alternatively, if a company fears the complexity of a major digital transformation and delays investment then they may find that entirely new market entrants steal their customers by offering a more customer-centric product.

There are some key areas of focus that should be analysed before commencing on a digital transformation project, both to mitigate against failure and also to increase the chance of success. New research published in Information Age highlights six key factors for success in digital transformation projects:

1. Leadership; is the company leadership really supporting change?
2. People; does your team have the skills you need?
3. Agility; are you able to change plan during the transformation?
4. Business Integration; how will the transformed business connect to the existing processes?
5. Ecosystem; what support do you have from suppliers and others in the value chain?
6. Value From Data; are you capturing the right data and analysing it at the right time?

These may appear to be obvious points that any executive team would consider before a major change, but it’s worth studying each factor in more detail because digital transformation projects do fail. Often the reasons for failure are clear – a lack of agility is a classic example. If your project is so large that it may take several years to implement then it is almost certain that the requirements in a couple of years will be different to now. Therefore agility is essential.

The increasing use of Artificial Intelligence is a good example why these six key factors are important. In the Information Age research 68% of respondents said that they had already had a positive experience of AI systems and 61% expected AI to be creating new jobs in the near future.

That 68% figure is quite high for a technology that is often talked about as a trend for the future. Technologies such as AI are becoming very important in the present-day business environment and many of these emerging technologies will lead to a fundamental change in business models and the competitive landscape. But digital transformation is not just about the integration of emerging technologies into your existing business processes, it is the enabling of new business models or services through the use of technology.

These six factors identified by Information Age really do speak to the way that a digital transformation project should be approached. Kodak was researching digital photography and yet they never saw Instagram on the horizon – digital transformation can completely change entire industries in a short period of time so this is an area of strategy that is essential to get right.

Robotics And Cognitive Technologies Change Your Business Forever

IBA Group
Mark Hillary

The growth in corporate robotics feels rather like an overnight trend, but automation using robots has been changing the manufacturing industry for at least three decades. The difference today is that robotics is no longer restricted to factory production lines. Automation today is far more advanced than a machine capable of spray-painting car parts.
The reality in today’s environment is that several technologies are blending together to create new possibilities and solutions. Robotics, machine learning, and Artificial Intelligence are naturally connected because automation no longer has to just be the simple repetition of programmed bots – we can now ask the system to learn how to get better.
The IBM Watson system is a great example of this. Watson is capable of reading 800 million pages of data a second. This capacity to absorb new information constantly makes it incredibly useful for complex environments that are constantly changing. Cancer diagnosis is a good example because a traditional doctor will train for many years and then will work with patients in a hospital so their capability to absorb new research is limited. By training real doctors to work with AI systems such as Watson we can support and enhance them – allowing doctors to access a second opinion that includes knowledge of all published research.
Softbank in Japan has connected their Pepper ‘general purpose’ robot to a Watson ‘brain’ creating the possibility for intelligent assistants that actually have a physical form. It’s easy to imagine nurses treating patients with Pepper offering additional advice, or a bank advisor explaining a mortgage to a potential customer and Pepper offering further information and automatically checking compliance to legal regulations.
But this convergence of technologies is not taking place at the same speed in every company, or even in every industry. EWeek magazine recently summarised five important trends that give a good oversight on the growing importance of robotics in industry today:

  1. Most companies are not yet using Robotic Process Automation (RPA), but are noticing those that are using it; Capgemini research suggests that 39% of companies are already using RPA and many are talking of extremely positive results – such as a reduction in repetitive work and an improvement in quality. The companies that have not yet tried RPA are noticing these reports and will move quickly.
  2. RPA works best when used to create a Centre Of Excellence (COE); RPA requires a cultural change so it helps to create a mindset that you are not just automating existing tasks, rather the plan is to improve how the company works.
  3. Once companies explore RPA they deploy it everywhere; companies that have piloted RPA initiatives find that it is not just useful in the back office – automation can be deployed everywhere.
  4. Human jobs are changed, not eliminated; as with the Pepper examples, in most cases RPA enhances and improves what humans can do rather than just eliminating their roles. In research published by McKinsey, they estimated that around 90% of work functions cannot be automated 100% – the role of automation is to increase quality and productivity, not eliminate humans from the workplace.
  5. RPA plus AI will lead to new cognitive opportunities; by created automated systems that can learn we are entering a new cognitive era of business. Research by OpusCapita suggests that 81% of executives believe that this combination of RPA with AI will significantly change their business inside the next 5 years.

This highlights two extremely important – and opposing – points. Executives mostly (81%) believe that automation and AI is about to dramatically change their business, perhaps even their entire business model. However, only a minority (39%) of companies have already launched an RPA project.
It’s clear that this is where the future lies for companies across all industries so the future seems bright for service companies with expertise in both these areas. I even think that the 5-year time horizon is rather long – in my opinion this will all change before 2020. RPA and cognitive systems are about to change your business forever – are you exploring the possibilities today?

Genetic Programming

IBA Group
Pavel Charnysh

Artificial intelligence, intelligent self-learning machines, systems that can advise on how to do work better, and robotics – all of this has always been like magic to me. When I studied at the university, I was carried away by these topics. It is even more fascinating to use knowledge from one field for another and thus solve the tasks that seemed unsolvable.

What do you think about the interaction of artificial intelligence and the theory of evolution, one of the most interesting open issues in biology? As I worked with algorithms and not hardware, I kept wondering, how we can teach a computer to be intelligent. I did research on the topic within an internal project at IBA.  This article gives an overview of Genetic Programming and my speculations on how to use it in software development.

As Wiki says, ‘Genetic Programming (GP) is an evolutionary algorithm-based methodology inspired by biological evolution to find computer programs that perform a user-defined task. Essentially GP is a set of instructions and a fitness function to measure how well a computer has performed a task. It is a specialization of genetic algorithms (GA) where each individual is a computer program. It is also a machine learning technique used to optimize a population of computer programs according to a fitness landscape determined by a program’s ability to perform a given computational task’.

In my view, GP is a way to find a solution using atomic user-defined blocks.

The following are the main principles of GP:

1. Initially, we are given past performance data to build the most suitable program for reproducing the same principles of data structure in the future. We assume that the initial data were produced by a kind of a black box. We have historical input and output data to reproduce this system with the same rules and principles for future use.

2. All programs are members of a population. It means that we get not the only solution, but a set of solutions and can choose the most suitable one.

3. Changes in a population are made using an iterative method. At each iteration, the programs that are most fit for crossover and replenishment of the population are selected.

4. A fitness function is used to determine, if a program is fit for the purpose. It is a user-defined metric that numerically presents the ability of a program to solve the defined task (to fit the mapping of the input and output parameters for a given data set).

5. The fittest individuals of a population are selected to develop the population just the way evolution selects species.

6. Changes that can be implemented in the surviving members are similar to biological evolution. A member can be mutated or crossed with another member of the population.

7. A stop condition is defined for a fitness function, when one can stop GP and pick up a solution.

To grow up a population, a programmer must define primitive blocks which will form an individual. These are terminals, including constants and variables, and primitive expressions such as  +, -, *, /, cos, sin, if-else, foreach, and other predicates. Any program can be presented as a tree built up of these blocks. This way, any individual of a population can be presented.

Genetic Formula
In this case, mutation and crossover are represented as following.

Genetic Figure 1
Genetic Figure 2

We just take a randomly selected node from one individual and use it to replace a randomly selected subtree of another individual. That is a crossover. We can also take a randomly selected node of an individual and replace it with a randomly generated subtree. That is mutation.

Genetic Programming is used for neural network learning and numeric computing, as well to approximate complex functions. I was researching GP to imitate activity of a definite person. An employee while doing his or her job can make both optimal and non-optimal decisions, which makes human thinking different from digital technologies. When you are asked to point to the south, you won’t be able to do it without a mistake, and this ‘white noise’ is our individual quality. Sometimes, we do not need an accurate answer to solve a problem. After gathering the data, we can imitate this employee’s behavior for new tasks using a computer program.

Genetic programming is also useful when creating an artificial player for a game with different difficulty levels. The behavioral algorithms of an artificial intelligent player are typically ideal and therefore a human cannot beat it, if it didn’t play at give-away. Consequently, we need to make the artificial intelligence a little bit human, allowing it to make mistakes from time to time. To this end, the GP algorithms are in place because they teach the artificial intelligence human behavior, including the ability to make mistakes.

 

Isn’t that remarkable? I think it is real magic and those who create smart computers are magicians. Who knows, maybe it’s a way put a soul into a computer.