Generative artificial intelligence (AI) has captured the imagination of organizations across industries, promising to revolutionize workflows and drive innovation. As public sector entities explore this transformative technology, a critical challenge emerges: identifying and prioritizing high-value use cases that align with specific business objectives and deliver measurable results.
In this article, we present a Amazon Web Services (AWS) framework to help public sector organizations adopt generative AI and unlock its true potential. By following a systematic process anchored in business strategy and value mapping, teams can prioritize high-impact use cases, align stakeholders, and measure the tangible benefits of their generative AI initiatives.
In addition to matching generative AI development priorities to business needs, executives and technologists must sufficiently understand the capabilities of generative AI in order to validate whether it is the right tool for the work. According to a study carried out by researchers at Harvard Business SchoolGenerative AI can increase the performance of highly skilled workers by 40% on tasks well suited to this technology. Conversely, when generative AI is used outside of current limits, its use reduces the performance of highly skilled workers by 19% on average.
Without a clear understanding of the technology and how to measure the business value of the proposed solution, public sector organizations may find the ROI of their innovation limited or unidentifiable.
Successful AI adoption follows many different paths, each unique to the organization’s mission and requirements. In thousands of customers, we have identified a common sequence of events that we call the AWS Cloud Adoption Framework for Artificial Intelligence, Machine Learning, and Generative AI (AWS CAF-AI). The following diagram shows the Artificial Intelligence Cloud Transformation Value Chain.
Figure 1. AWS CAF-AI transformation value chain diagram.
The AI transformation steps are as follows:
- Work backwards from your understanding of what AI allows you to do.
- Define what your expected business outcomes are over time.
- Build the transformation your business needs to go through.
- Develop the fundamental abilities that enable this journey.
In the following sections, we take a more in-depth look at each of these areas through the lens of AnyOrganization, a public sector organization active in financial regulation.
Work backwards from your understanding of what AI allows you to do
AnyOrganization’s Chief Operating Officer (COO) recently read an AWS blog post which dealt with generative AI for the public sector. The COO believes generative AI has the potential to address some of his organization’s biggest challenges in customer experience, process improvement and employee productivity. After consulting with his team, the COO prepares a recommendation for a proof of concept (POC) project aligned with one of his existing objectives and key results (OKRs): improving the agency’s exam coverage.
Define what your expected business outcomes are over time
The COO knows that for this POC to be successful, he must accurately forecast the anticipated business value and how he will achieve it. Their OKR, which improves the agency’s review coverage, stems from an inherent limitation of the human reviewer. Given their limited skilled manpower, reviewers can only thoroughly review a small percentage of the documentation. Recent events in the financial markets have challenged AnyOrganization to do more with fewer resources.
After consulting with the AWS account team, IT, and business stakeholders, the COO approves the following expected business outcomes after implementing the new generative AI solution.
- Increase the percentage of documentation reviewed from 20% to 100% with generative AI, while maintaining human review at the current level of 20%.
- Effective use of AI generative documentation prescreening allows humans to review the most relevant 20% of documents, resulting in increased reviewer job satisfaction and 50% more results requiring review of a higher level.
Build the transformation your business needs to go through
AnyOrganization’s COO reviewed AWS best practices and found that an organization’s ability to derive measurable business value from AI-driven innovation starts with ensuring the following four areas of transformation are in place.
Transformational domains
Technology – Do your development teams have access to the necessary AI and machine learning (ML) tools and services? Are your existing service activation procedures ready to evaluate, approve and protect these powerful tools?
Process – What long-standing organizational processes need to evolve to optimally use new technologies? Are existing data management practices sufficient to fuel your AI/ML flywheel?
Organization – How are your business and technology teams orchestrating their efforts to create customer value and meet your AI-driven strategic intent? Do your legal and compliance teams need to integrate more closely with development?
Product – How can you reimagine your business model to create new value propositions (products, services) and new revenue models that capitalize on the capabilities of AI? Where will you allocate the new capacity freed up by service improvements?
Transforming these areas and enabling them to use AI depends on your core capabilities across business, people, governance, platform, security and operations.
Fundamental abilities that enable this journey
The AWS CAF provides six perspectives to visualize the capabilities required for successful AI adoption.
Business – This perspective helps ensure your AI investments accelerate your digital transformation and AI ambitions and business outcomes. In particular, we explain how to put AI center stage, reduce risks and increase customer outcomes, thereby enabling the formulation of an AI strategy.
People – This perspective serves as a bridge between AI technology and business and aims to evolve a culture of continuous growth and learning, where change becomes the norm. AWS offers many options to improve your knowledge and that of your teams in generative AI. AWS Skills Builder offers a variety of free and on-demand options for generative AI. Of particular interest to readers of this article, our Generative AI Learning Plan for Decision Makers.
Governance – This perspective helps you orchestrate your AI initiatives while maximizing organizational benefits and minimizing transformation risks. We address the changing nature of risk and, therefore, the cost associated with developing and scaling AI. Additionally, we are introducing a new AWS CAF-AI feature in this perspective: responsible use of AI.
Platform – This perspective helps you build a scalable, enterprise-grade cloud platform that lets you leverage AI-based or AI-infused services and products and develop new custom AI solutions. We illustrate how AI development is different from traditional development tasks and how practitioners can adapt to this change.
Security – This perspective helps you ensure the confidentiality, integrity, and availability of your data and cloud workloads. We extend our existing security guidance from this perspective to show how you can reason about and respond to attack vectors affecting AI systems through the cloud.
Operations – This perspective helps ensure that your cloud services, particularly your AI workloads, are delivered at a level that meets your business needs. We provide guidance on how to manage AI operational workloads, keep them up and running, and ensure reliable value creation.
Your AWS account team can help you organize structured assessments of your current capabilities through mechanisms such as AWS Cloud Maturity Assessment (CMA), Experience-Based Acceleration (EBA)or a Executive Briefing Center (EBC) session focused on developing your generative AI strategy.
Further reading