AI for Business
95% of CTOs, CIOs, and IT directors believe that AI will be the most significant driver of innovation across almost every sector in the next one to five years. This is one of the insights from a survey conducted by IEEE and it speaks to the growing importance of AI. The conversation about this technology spans possibilities, sectors, and use cases, both present and predicted. Depending on where you stand, this might seem very scary or exciting, or both.
In business today, AI is being used beyond IT departments, through integration in key parts of an organization’s operations. A good rule for the extent of AI adoption is that anything that is a process will eventually be done by AI. It also provides value through analytics and engaging with people. The technology’s applications include machine learning, automation, activity recognition, robotics, activity recognition, and natural language processing. It is able to do this through a system that is built using historical data and defined by set parameters.
AI does not work for every single part of the business process, only for the ones where it makes sense; because they are repetitive or can be done more effectively by AI.
It is presumptive to have jumped into this without a simple definition of AI, and that will be remedied now. For the purpose of this piece, AI is any machine that does things the brain can do (Financial Times).
There are different views on the extent to which businesses can and should use AI. Because people make those arguments, a natural follow-up to it is what AI means for a human workforce. Different statistics paint a picture of what employment will look like as AI use increases. According to pWc 7 million jobs will be replaced by AI between 2017 and 2037 in the United Kingdom. Its data also says that in that period, 7.2 million jobs will be created. The World Economic Forum’s Future of Jobs Report also shows a similar prediction: it states that AI will replace 85 million jobs globally by 2025, while creating 97 million new ones in that time. The predictions share a trend- AI is not expected to wipe out employment, rather it will reconfigure it. It will make certain jobs pointless for humans to carry out while creating new jobs in response to the different needs that will arise as adoption increases and systems become more complex.
Predictions can be wrong but they give us something to work with. In this instance, it allows organizations to consider where they find themselves in the grand scheme of things; what it means for their people, and how they can leverage this technology to improve performance, without doing harm to their people. This, therefore, highlights four areas to consider:
- Reskilling and Training
- Adoption and Application
- Human Capital Management
- Research and Innovation
Reskilling and Training
As the nature and structure of labor and labor sharing change in an organization, today’s workers will need to be trained for a future that is already here. This is necessary from both an organization, individual, and system perspective. Organizations are usually responsible for training and development, in order to improve their human capital and maximize productivity. The specific contexts and needs of different organizations means that they are the best placed to train their people in line with said context and needs. Training and development for AI adoption involves two classes of workers; those that will work directly on AI systems, and those whose jobs will be reconfigured as they work with AI or give up certain aspects to it. For the former, training is important in order to improve on skills that should already be present, while the focus for the latter is on introducing them to the technology, guiding their interactions with it, and staying updated on changes in application.
At the individual level, workers cannot depend solely on organizations or schools to train them. AI in business is one of the features of our knowledge based economy, and it is one in which learning will be increasingly blended in terms of learning platforms and skills being acquired. Individuals who do not want to be left behind will have to take responsibility for their own learning, especially in terms of gaps, the needs of organizations, and their own projections for career growth. The importance of education systems cannot be overemphasized, and the calls to improve those systems are increasing, and for good reason. However, it is unlikely that disparate education systems across the globe will catch up to technology, at least not in the sense of traditional instruction. Our understanding of education systems should therefore change, to include the role of organizations and individuals, in addition to traditional schooling.
AI does not only necessitate training and development, it is also the foundation on which an important mechanism of its delivery is built; personalized learning. AI’s functionality in this regards includes curating and recommending knowledge for learners, learnbots being used for corporate training and onboarding, analytics which can be used to improve training as well as identify needs in the organization, introducing learning into the workflow, and learning experience platforms which improve the learning experience.
Adoption and Application
Adoption cannot take place in the same way across or even within sectors. AI is relevant for all kinds of businesses, but its adoption and application will usually be determined by factors such as capability, infrastructure, capacity, and even the clients (in client-facing applications). . The promise of AI is that its various applications can serve a range of organizations needs, in line with the value that they seek to deliver to customers. Three factors that should guide adoption and application are sector, market and business value.
Some sectors will by nature be more reliant on AI than others, depending on whether it is a core part of its products and services, or if it supports certain business functions. The leaders in AI adoption by sector are currently telecom, tech, and financial services.
After sector comes market. Markets are different, and companies in the same sector must take into consideration the markets they serve. Some are more advanced than others, and the rules they adhere to are not the same. Generalization does not work, for example, assuming AI use in pharmaceuticals in Germany can be applied exactly the same way in pharmaceuticals in Bangladesh. Adoption must make sense, and be suited not just to the sector, but the market being served. This leads to the final factor.
The value discipline(s) of an organization will have a key role to play in terms of AI adoption and the applications used. Whether it is pursuing customer intimacy, product leadership, or operational excellence, an organization can and will find value in a number of AI applications. In terms of customer intimacy for example, chatbots and sentiment analysis are being used to improve customer engagement. Personalized customer experiences are also being provided through AI, which uses customer data to keep track of preferences and make person-specific recommendations.
Find out more about adoption and application: Reimagining Your Business for AI
Human Capital Management
The illustration above shows the shift in division of labor between man and machine by contrasting two years(2018 and 2022). In more complex tasks such as reasoning and decision making and communicating and interacting, the share of manhours remains much larger than machine hours. Tasks such as information and data processing as well as looking for and receiving job related information are showing machines taking up a significant share of work time. This is because they are routine processes that are not complex or driven by creativity.
It brings us back to how work configurations will change; in terms of what tasks there are to be performed, and how they will be performed. The demands on manpower will be different and organizations will have to take this into account, through the entire employee lifecycle, from how they attract talent to recruitment, onboarding, talent development, retention, and separation. Similar to reskilling, AI tools can also be used to manage some of the changes that come about as a result of adopting the technology. They cover different aspects of the employee lifecycle, as well as key aspects of human capital management (HCM). AI for example is used in the recruitment process in order to source and select candidates, with the aim of having a more objective system, and reducing the burden on recruiters. It is also being used to improve human resources service delivery in organizations through chatbots which can carry out some of the more routine interactions that happen in a workplace.
It is recommended that organizations looking to adopt AI in their HCM consider their specific needs and the systems required to support getting maximum value out of adoption.
“Don’t invest in it, if you and/or your organization is not fully prepared to adopt it. More specifically, implementing an AI strategy for HCM is easy. All it takes is time and money. However, nurturing the business systems and data elements that allow AI to aggregate and exploit better outcomes i.e., do what it was intended to do, that’s a completely different matter. In the context of HCM, organizations should have consensus on the fixed labor elements such as budgets, task standards, averaged cost, workload factors and production drivers, all of which are collapsed under AI optimization…”- Scott Morgan, Infor.
Research and Innovation
Adoption won’t be the same across industries, but it will happen. Whatever sector your business is in, and whatever market it serves, it must pay attention to key changes happening in the world, including AI sponsored change.
Hans Christian Boos says that AI is every company’s business and companies that do not focus on building or consolidating on a strong brand, innovation, and service will not survive in the coming years. Its ability to integrate itself across the different areas of a business model makes AI the concern of any organization looking to lead. It will matter how it is being used, who is using it best, and who provides the technology. The last one is especially important.
As pervasive as AI is across business and daily life, its development is dominated by a few key players. The concept of AI-powered monopolies has been put forward, and its meaning is just as the title infers. They are or will be technology companies who, by reason of their dominance in two essential inputs for AI (data and computing power), will become even more powerful than before as more aspects of human life come to rely on AI systems.
The concept of a technology company should be thought of in the encompassing sense of one in which key parts of its business model leverages technology. In this sense, we come to understand that 21st century organizations should be tech companies, because technology has become essential across all components of an operation model. It would be unwise to neglect understanding AI as a key technology for business today, whether as a user or a provider of its services.
The research and innovation efforts of organizations should therefore be guided by this, in order to discover how best to utilize the opportunities of AI, the extent to which it can and should do so, as well as creating mechanisms for measuring progress and addressing challenges.
“AI is good”. “AI is bad.” Those are two narrow perspectives that assume mutual exclusivity of AI’s nature. AI is largely what its developers, users and policy actors make it. In the context of business, its use has been defined as a technology that can and should replace human labor or one that should not be used, in light of the first definition. Both definitions are problematic, when we consider the capabilities of AI and the human resource needs of organizations.
AI is a tool, not a person, at least not yet. Tools are created to improve the results of human efforts, and how well or badly they do so, again, depends on us. The technology is not going anywhere, and like other exponential technologies, it adheres to Moore’s law. As it becomes more complex, it will increase in efficiency while reducing in cost, making it more accessible to users. Intellectual energy would therefore be better concerned with the process of adoption and how it can be done to create the best outcomes for organizations and people.