Anyone who has browsed Amazon, YouTube or Google recently might be totally unaware of the amount of artificial intelligence (AI) that they have just used, as they decide to pursue a recommended video or look at a recommended item to purchase. In these scenarios, the algorithms used to assess your past activity and influence or predict your future actions have blended into the background to become part of the experience. The same goes for consumers who use smart thermostats and alarm systems in their homes. These devices observe behavior, and build an understanding of the user, allowing for an improved experience with fewer errors or failures.
Key Fact #1: AI is already here. It is becoming cheaper and more available daily.
In truth, these technological achievements aren’t exclusive to AI. They are a hybrid of techniques whose abilities cross the boundaries to collectively serve their owners. These other techniques are machine learning (ML), which makes the observations and corrects actions accordingly, and the internet of things (IOT), which helps ensure communication between devices over the internet – something that up until now was restricted only to computers and smartphones.
Is It Intelligence or Just Fast Computing?
The term intelligence might be subject to debate and is influential in both the acceptance of and resistance to progress in this area. The true basis of AI is the capacity for high-speed, in depth computation. This is what makes facial recognition a reality, for example. Although humans are far better at recognizing a face in a crowd or inferring an emotion, AI-based facial recognition programs can compare millions of existing facial records in milliseconds, even accounting for variation in light and angle.
AI is not a single thing. It can be based on a combination of learning techniques such as:
- Supervised learning, in which machine learning is guided by data such as a database of phonemes that would help a text-to-voice application learn how to better synthesize human speech.
- Unsupervised learning, which lets computers detect patterns to build its own database of facts.
- Reinforcement learning, which uses rewards and punishments, to help a computerized system learn essentially through trial and error.
Recently an AI-based art algorithm created a portrait, entitled the Edmond de Belamy, that it designed after observing over 15,000 portraits painted by humans over the past 600 years. It sold at Christies for almost half a million US dollars.
On a more pragmatic front, AI algorithms are being used for purposes such as predicting patient recovery based on medication interactions, family histories, and geographic location. Millions of AI applications already exist within a range of industries.
Jobs Are Not Disappearing, But They Are Changing
Even in the absence of true intelligence, AI is already changing the business landscape. Travel agents, for example, represent one line of business that has already been substantially impacted by AI-based travel sites that have access to far more real-time data about destinations, hotels and flights, and which can tailor the options based on data about each customer that it has already obtained from the customer or from a third party.
Key Fact #2: AI doesn’t eliminate jobs, but it does change them.
Jobs Are Not Really Disappearing
Jobs of all types will follow suit. This might mean the elimination of certain types of jobs, but that is not the same as destroying jobs. For example, many food store/supermarket chains now use self-checkout terminals, which use AI to help individual shoppers scan their groceries and cross-check the weight of these items against inventory and pricing databases. These devices appear to threaten the livelihoods of cashiers, however, each zone in which a self-checkout device is used requires human beings to take care of security (anti-shoplifting), helping customers with checkout problems, unjamming or overriding processing problems, and, further up the supply chain, designing, programming, delivering, installing, and maintaining the devices.
The jobs are not disappearing, but they are changing. Some will require additional training, but this has been the case in any industry for hundreds of years.
Jobs are Modularizing
AI also allows companies to modularize their hiring or team-management processes according to more accurate assessments of skills and availabilities. A new trend among large companies is to extend the concept of remote work by individuals to the use of remote teams. This consists of individuals, either employees or freelancers, who are pulled together as needed to work on a specific project for as long (or as short) as needed.
This dynamic team-based approach forms part of what is called the gig economy, which is best maintained and optimized through an AI interface that calculates the availability and productivity potential of freelancers as individuals and as part of the team. It is not new per se. Contractors and vendors have always been part of any economy. But it is the dynamic and highly mobile way in which it is now done that is different.
This changes the dynamic of hiring and employee retention, potentially reducing fixed payroll and benefit costs while ensuring access to qualified teams at short notice and on an as-needed basis.
The Rise of Predictive Analytics
One of the most beneficial outcomes of AI is the capacity for proactive services through predictive analytics. Sensors connected to the Internet of Things, paired with machine learning, allow for refined activities such as improved routing of trucks and ships, deployment of timely service calls for machinery, and the ordering and stocking of goods based on customer traffic and numerous other influencers. It is also better able to anticipate undesired events such as machine failures.
Key Fact #3: AI, paired with predictive analytics can change a customer’s future.
Predictive analytics helps move entire industries from reactive to proactive, ensuring that customers’ needs are met before any gap in service occurs.
How Can Organizations Embrace AI?
It is important that organizations avoid considering AI as one single event. It is equally important that the related concepts, machine learning, the Internet of Things, and predictive analytics are understood in the context of their individual abilities, but also in how they work together. Deployment of these techniques is ultimately more successful when approached as a continuous process.
Leaders and decision makers need to be able to observe their processes and their customers’ needs through the prism of innovation: How could we do this better? Where are the gaps? What are our customers doing with their time, actions and money? This, then needs to be applied to the second question – what can AI and related technologies do to improve things?
Currently, industry acceptance of AI is comparatively low, being in its early stages. Leaders and teams struggle to envision just how these technologies might fit into the status quo. In many situations, the business case is still not there – not enough information or vision is available to help paint a picture of how AI would service a business profitably in the coming years.
There is also a skills gap. Not enough people are sufficiently skilled in AI technology to reliably deliver and maintain a platform in-house.
Thus, awareness and training are key. Although AI itself is computer-based, it needs human beings in decision-making and deployment positions to ensure the system becomes a viable, going concern.
Enhanced Cognitive Skills Are Still Needed
In assessing the future of work over the coming 15 years, numerous think tanks point to an interesting parallel to AI, being the need for enhanced social and intellectual skills in the workforce. As the need for basic mechanical and functional skills is answered by intelligent automation, there will be a heightened demand for higher level cognitive abilities such as critical thinking, entrepreneurial skills and emotional intelligence.
As the intelligence of our computing machinery grows, so too will the intellectual capacities of the humans working alongside them. For example, an IT specialist who writes code has traditionally not been expected to excel in negotiation skills or organizational intelligence. A coding position has traditionally been seen as an engineering role, dealing with data, functionality and construction in a very objective manner. Yet as IT experts get called to the table to discuss security or strategy with senior management, their abilities in code writing must now be paired with the ability to discuss and influence at a much higher level.
This turns into a responsibility for companies to ensure they are delivering the types of education that will groom employees for these positions.
Building an AI platform ultimately demands that companies recognize the value of data as a dynamic, and not static, element of business, and that they set their companies and their people up to make the best use of it.
- Artificial intelligence is a combination of resources including high speed data processing, Internet of Things, machine learning, and predictive analytics
- Jobs will change to address heightened technological and social skills
- Embrace AI by observing customer need and questioning how it can be addressed.