AI and chatbots: Driving the future of government IT service management
- By Marcel Shaw
- Dec 15, 2017
Artificial intelligence is driving how government agencies will serve the public and their internal workforce in the future. Social services agencies, public safety organizations, the IRS -- any agency tasked with serving millions of citizens -- can benefit from the efficiency and customer satisfaction that AI can provide. Similarly, the government workforce can benefit from more up-to-date and accurate information supplied by AI on items such as policy changes or new directives.
How exactly AI fits into the agency of the future is a topic gaining steam as technologists and analysts ponder how AI will be evolving and expanding to more sectors where it can enhance service. In its October report, the Information Technology Industry Council detailed principles for the secure use of AI and encouraged public-private partnerships for investment, innovation, scalability and education of the workforce of the future. The ITI estimated that by 2020, the U.S. market for AI technologies that analyze unstructured data is projected to reach $40 billion, potentially generating more than $60 billion in annual productivity improvements.
In Gartner's “Is Your Digital Government Platform Ready for Virtual Assistants and Chatbots?” report, the author said that “chatbot and conversational AI platforms are opening new government service delivery channels. Government CIOs need to quickly determine the role of these channels, adjust their digital service delivery strategies and extend their digital government platform to exploit these new opportunities.”
AI implications for IT service management
For ITSM, it’s useful to think of AI and chatbots as augmented intelligence, or as the ITI calls it, human augmentation. We’re not looking at a sci-fi dystopia where evil robots replace humans; this is AI helping people do a better job with ITSM, with serving internal and external customers.
There are three key areas in which AI will be causing positive, disruptive changes in how ITSM functions:
1. Point of entry. AI-driven chatbots will enable automated ITSM solutions to interpret incidents and requests more accurately. They will learn different scenarios and route requests to the correct back-end process. At the very first touchpoint, AI will provide a more extensive decision path, thereby directing the inquiry to the appropriate solution with fewer mistakes.
AI-powered chatbots can learn who service requesters are, what they know and what they don’t know. This profile will help route inquirers who frequently contact a particular agency to the correct solution, saving valuable ITSM time. Government employees can get better training suggestions when AI understands their knowledge gaps.
2. Automated back-end processes. ITSM consists of back-end processes designed to manage any request or issue entered into the system. The power of AI is that it can recognize patterns and accommodate an infinite number of resolution scenarios, far more than is possible by humans. AI will collect patterns of problems coming in and, with integrated technology solutions, enable automated troubleshooting to resolve the problem.
For example, integrating an ITSM solution with one that provides IT operations analytics would enable an AI-powered ITSM solution to be notified of potential network issues. If it is further integrated with a security solution, it could detect unusual activity such as excessive browser crashes in one day. The ITSM solution would be able to investigate that issue and cross reference the data with the IT security solution to find any patterns that might explain the anomaly. When the ITSM solution logs the “problem,” it would be able to provide predictions about how the problem will progress and recommendations about how to fix it. AI will remember past experiences, which will help determine whether browser crashes are the result of an internal network issue or caused by a possible cybersecurity attack.
3. Knowledge management. AI solutions that have access to trusted knowledge databases will, over time, learn patterns that exist within an organization to help it more efficiently solve problems. They will not only provide answers to IT questions but also be able to provide training and tips for end-users and analysts. They will efficiently and accurately updating the knowledge database with more current and relevant documents based on new scenarios and new fixes.
Getting from now to AI
Successful integration of AI into government agencies will also save money by reducing costs and increasing service desk and service management efficiencies. An April 2017 Deloitte Insights report, "AI-Augmented Government," conservatively estimates that automating tasks that computers already routinely do could free up 96.7 million federal government working hours annually, potentially saving $3.3 billion. Some of these tasks include reducing service request backlogs, pushing vital information to employees via messaging apps and helping agencies quickly cull through big data to identify trends and speed resolution.
All well and good, so how do we get there?
Connections. To facilitate AI integration, agencies need an automated infrastructure that connects all their technology silos. Merging these technology operations will create the foundation for AI.
Integration. This connected, automated infrastructure should integrate ITSM, operation analytics, security, procurement solutions and contracting solutions to effectively collect new information, identify a solution and send data back through ITSM to provide a satisfying response to an inquiry or current information on events to internal customers.
Business process management. Planning for augmented intelligence necessitates a different approach to the workforce as well. Agencies must train employees to manage chatbots and to become more technical and diverse in their knowledge of agency operations. Since connectivity is essential to AI, and silos must be integrated, an agency's workforce must also mirror this connectivity of knowledge. All ITSM professionals should understand more about how security works, for example, to help identify threats. The jobs of the future will require managing automated tasks across departments as AI becomes a standard part of business processes.
Response times measured in days are no longer acceptable. Government consumers today -- both the public and agency employees -- expect results in seconds. Investing in augmented intelligence will pay off in efficiency, cost savings and improved service. New jobs will be created as AI gains traction in the government sector. Starting to build an AI-enabled infrastructure now will reap benefits in workforce satisfaction and ITSM effectiveness.
Marcel Shaw is principal federal solutions architect at Ivanti.