There are few technologies receiving as much attention today as artificial intelligence (AI) and machine learning. These technologies hold incredible disruptive potential—revolutionizing how we interact with machines, enabling automation, and driving productivity to previously unthinkable levels.
It’s this potential that is driving research and innovation in AI, and delivering a new and exciting generation of AI technologies to the market. This potential has also spurred exponential growth in both the capability and number of AI solutions and services being implemented across the commercial and federal sectors.
In fact, many industry analysts see AI as a market with the potential for rapid and sustained growth in the near- and long-term future, with analyst firm, IDC, claiming that, “artificial intelligence (AI) systems are forecast to reach $57.6 billion in 2021.”
With so much excitement and fervor around AI, it’s understandable to perceive it as something extremely new. Although today’s new AI solutions are, in fact, revolutionary, that’s not completely accurate. In truth, AI has been around nearly as long as electronic computing itself.
Alan Turing—whom some of our readers may know from the popular movie about his life, The Imitation Game, where he was played by Benedict Cumberbatch—pioneered both electronic computing AND artificial intelligence in the 1940s.
Since Turing’s work on the general-purpose computer, and the articulation of the Turing test, AI has come and gone from the public eye. But it was never forgotten and it was never abandoned. Instead, it’s grown and matured in the form of expert systems, natural language processing, image recognition, speech recognition, robotics, machine learning, data mining, and other implementations that have been utilized in both the public and private sectors.
If there’s a reason why AI feels different today, it’s probably due to its pervasiveness. AI solutions aren’t just being used in research settings anymore—they’re inside our homes in the form of devices like Siri, Alexa and Google Home. They’re also accomplishing impressive things in both the laboratory and office. And this pervasiveness is a result of three factors:
- The rise of Big Data: Data is the lifeblood of AI. AI and machine learning solutions need data from which to learn, and detailed data—with metadata—is widely available today. Since 2010 vast amounts of training data have become available, including image data, social media data, machine logs, mobile geo-location data, wearables and many more. All of these sources of data can be valuable for training AI algorithms, and they’re extremely vast.
- Economical computing: Developing AI solutions is no longer restricted to large technology companies and other organizations with massive budgets. The large amount of storage and compute resources required for AI processing is available to even small departments and start-ups thanks in large part to the cloud. Cloud solutions provide elastic and scalable compute and storage infrastructure with little capital investment upfront, making it easier for companies of all sizes to develop AI solutions. Readily available software development resources with AI APIs, cognitive services, ML tools, and data science platforms also level the playing field and make it so that developers don’t have to be data scientists to create AI solutions.
- Open-source algorithms: The emergence of distributed data processing platforms such as Apache Hadoop and Spark, and Deep Learning frameworks like TensorFlow have given rise to an ecosystem of open-source data analytics and machine-learning tools. These are increasingly powerful open-source tools that contribute to the development of new AI solutions available for anyone to use.
Although AI isn’t new, it is experiencing a renaissance thanks in large part to Big Data, economical and powerful compute and storage, and open source data science algorithms. As a result, new, exciting technologies are entering the market and are disrupting how we work and live. It’s for this reason that AI is one of the most exciting new technologies today.
In fact, AI is so potentially transformative for the federal government that we’re going to be looking much closer at it over the course of the next month. In coming articles on Thinking Next, we’ll be speaking with cloud providers and emerging technology companies that are creating innovative new AI tools that can help federal agencies drive operational efficiency and better serve constituents. We’ll also be taking a deep dive into how AI can combat the challenges facing today’s federal agencies.
For additional information on AI and CSRA’s artificial intelligence and machine learning capabilities, click HERE.