Saurabh Prabhat is Conduent's Director of Analytics Practice.
Although Artificial Intelligence has been around for decades, we’re at a pivotal point for its impact on businesses, consumers and society. AI is both a catalyst and result of increasing consumer expectations for more personalized, ubiquitous and intelligent experiences in every area of their lives.
New AI-driven real world applications are being enabled at massive scale in almost every industry and domain. These solutions can sense, learn and interpret human behavior and complex situations. AI solutions are augmenting humans by assisting, making or even acting on critical decisions, transforming intelligence, perception and decision capabilities.
AI is an aggregation of many technologies
It’s important to understand that Artificial Intelligence is not just one technology, but rather an aggregation of many technologies often working together. At a very basic level, all AI solutions break down into a three-step series of actions:
- Sensing or seeing information/data
- Sense-making or interpreting that data
- Deciding and/or taking action
These three steps are followed by learning and adaptation by the system based on the outcomes. All these actions, and all steps in the process, could either be done by machines, humans or a combination of both.
AI platforms driving industry transformation
Just about every industry is being influenced by the intelligent, individualized and timely decision and process automation that is enabled by AI. The chart below shows just a few examples from a wide variety of industries.
Value Proposition enabled by AI Solutions
When people think about AI, their minds often drift to the idea of autonomous systems “running the world.” In some scenarios AI solutions could replace humans, but in most applications intelligent machines will augment human activities — with their ability to quickly process enormous amounts of data, for example. We should, therefore, consider a spectrum of applications and value propositions made possible by AI, as represented in the quadrant diagram with applications ranging from repetitive tasks to self-learning systems on one axis and tasks which can be fully completed by machines to those requiring a partnership between humans and machines on the other.
It’s important to note that there are many tasks that can be performed without any human intervention. These tasks, such as assigning support cases, distributing leads or routing calls based on rules can be fully automated — saving businesses and people valuable time and resources.
More complex tasks, which require a mix of skill, judgment and experience, tend to be better suited to a combined team of bots and people. Virtual assistants can auto-pilot planes after take-off, or self-driving cars can operate with or without a human in the driver's seat.
These man/machine relationships have evolved over time with new trust and comfort building such that we are increasingly collaborating with machines/robots like digital assistants and even delegating complete ownership of tasks.
These AI technologies-based offerings are growing rapidly. According to IDC, by 2019, the worldwide market for cognitive software platforms and applications will grow to about $16.5 billion.
Five key drivers of cross-industry AI growth
Understanding how AI is impacting industries is important. What’s equally important—and critical — is understanding why automation, machine learning, natural language processing and other elements of AI are advancing so quickly.
We see five primary drivers for this growth:
- Technology is evolving. AI applications and services are more advanced, more integrated and becoming more pervasive due to the easy availability of high-powered computing along with massive amounts of data and new advances in algorithms. AI solutions are becoming easier to develop with cloud capabilities including IaaS (Infrastructure as a Service) and PaaS (Platform as a Service), freeing up technical staff from maintaining infrastructure so that they can focus on strategic capabilities.
- Costs are being driven down. Though data creation, consumption and content transport are all increasing, costs for bandwidth, data collection sensors and data processing are rapidly decreasing. Companies are investing in algorithms and platforms to store, transport and process massive data for extracting valuable insights
- Digital transformation is becoming more pervasive. Companies are driving more of the business process functions to the digital world which in turn is helping to eliminate bottlenecks and provide context-specific access to data. This, in turn, speeds the adoption of AI solutions.
- Challenges with human processes. Issues with employee engagement and inefficiencies are also causing companies to invest more in AI-based solutions. AI helps to automate those tasks that humans may not want to do, or that may be more efficiently done by machines.
- Companies are making increased investments and expectations are growing for AI solution outcomes: As AI technology platform vendors, consulting firms and managed service providers implement solutions for their customers, they create repeatable processes which can be applied within a single industry, or across many sectors. Information management executives are learning about the benefits of AI from their peers and counterparts in their industry, and investing in their own environments.
Big plans for the future
Organizations today are excited about the potential opportunities that AI-based solutions can offer their business, enabling more individualized, immediate and intelligent digital interactions. Most have plans to increase investments, but many are still unclear about how to get started and how will they measure the impact of AI on their business.
At Conduent, we believe that it’s important for enterprises to address the gaps between AI strategy and execution, to enable more real and tangible outcomes. Companies that have a clear roadmap and leverage AI-based solutions get deeper insights into their business, so they can make better informed decisions and implement them faster.
About the AuthorMore Content by Saurabh Prabhat