Generative AI in the “Pit of Empty Promises”: Understanding the Wheelhouse Risk Cycle
The recent Wall Street Journal article, "The AI Revolution Is Already Losing Steam," provides a stark reality check on generative AI's limitations and challenges. This emerging technology, which once promised to revolutionize industries, now appears to be falling into what we at Wheelhouse Advisors term the "Pit of Empty Promises" within the Wheelhouse Risk Cycle.
What is the Wheelhouse Risk Cycle?
The Wheelhouse Risk Cycle, formerly the Digital Risk Cycle, was renamed to reflect its broader applicability beyond digital transformation and honor its creator, Wheelhouse Advisors. In today's rapidly evolving business landscape, organizations face risks across various domains, including technology, finance, operations, and compliance.
Source: Wheelhouse Advisors IRM Navigator™
Organizations can navigate these challenges more effectively by adopting a comprehensive risk management approach. The term "Wheelhouse" symbolizes a holistic and adaptable strategy, encompassing all risk areas within an organization's purview.
The Wheelhouse Risk Cycle is a strategic framework designed to help organizations anticipate, understand, and mitigate the risks associated with digital transformation. It extends beyond the Gartner Hype Cycle by focusing on the inherent risks at each stage of technology adoption.
Let's explore this concept through a series of questions and answers:
Q: What are the stages of the Wheelhouse Risk Cycle?
A: The Wheelhouse Risk Cycle consists of five key stages:
Risk Catalyst: The initial stage where potential risks are identified and monitored.
Pinnacle of Perils: Where risks are fully assessed and robust controls are implemented.
Pit of Empty Promises: Where the reality of the technology's limitations becomes apparent, and previously unforeseen risks materialize.
Incline of Integration: Where risks are integrated into broader business operations and aligned with regulatory requirements.
Mesa of Mitigation: Achieving a stable state where risks are continuously monitored and managed.
Source: Wheelhouse Advisors IRM Navigator™
Generative AI: Falling into the Pit of Empty Promises
According to the Wall Street Journal, the initial excitement and rapid improvements in generative AI, such as OpenAI's ChatGPT and Google's Gemini, have begun to plateau. Despite significant investments and high expectations, the technology's capabilities have shown only incremental gains recently. This scenario exemplifies the Wheelhouse Risk Cycle's Pit of Empty Promises stage.
Q: What are some key challenges facing generative AI, as highlighted in the article?
A: Several challenges are contributing to the disappointment in generative AI:
Slowing Pace of Innovation: The rapid advancements seen in the early days of generative AI are now slowing, with significant barriers to further improvement.
High Costs: The exorbitant costs of AI systems remain a significant barrier, with the industry spending far more on infrastructure than it generates in revenue.
Limited Usefulness: The real-world applications of AI are fewer than initially imagined, and the technology has not yet become the productivity booster it was touted to be.
Commoditization: AI technology is becoming more commoditized, reducing startups' competitive edge and profitability.
Native Enterprise AI Will Succeed as Legacy Software Struggles
Recently, C3.ai CEO Tom Siebel provided insights into why his AI software company thrives while traditional enterprise software companies struggle. Siebel highlights that legacy software companies, including those with legacy Governance, Risk, and Compliance (GRC) technology, are having trouble incorporating AI into their existing applications due to outdated technology stacks. These legacy systems were often built to manage regulatory compliance requirements and not the full spectrum of integrated risks modern enterprises face. In contrast, C3.ai's native enterprise AI applications are designed to meet current demands.
Siebel points out that CEO-level involvement in AI investment significantly drives AI adoption. He suggests that budgets for AI are flexible and often created on the fly by top executives who recognize the strategic importance of AI. This high interest indicates an increased risk appetite that can help push AI through the Pit of Empty Promises into the Incline of Integration. As AI moves through this phase, software companies that effectively incorporate AI into their application layers will be the ones that succeed.
Q: How will we know when generative AI begins to climb the Incline of Integration?
A: We will start to see signs that generative AI is climbing the Incline of Integration when the following occur:
Widespread Adoption and Integration: Generative AI tools and applications will be seamlessly integrated into everyday business operations, providing tangible improvements in efficiency and productivity.
Regulatory Alignment: Companies will align their use of generative AI with evolving laws and regulations, ensuring compliance and ethical use of the technology.
Demonstrated ROI: Organizations will start reporting measurable returns on investment (ROI) from their generative AI initiatives, justifying continued and expanded use of the technology.
Improved Capabilities: Incremental improvements will give way to significant advancements addressing current limitations, such as reducing operational costs and overcoming data limitations.
Enhanced User Training and Acceptance: Employees will become proficient in leveraging generative AI tools, supported by comprehensive training programs, and the technology will be widely accepted and trusted across the organization.
Navigating Digital Risks with IRM
Given these challenges, how can organizations effectively manage the risks associated with generative AI? The answer lies in leveraging Integrated Risk Management (IRM) technology within the Wheelhouse Risk Cycle framework.
Q: How can IRM help manage digital risks, especially in the context of generative AI?
A: IRM provides a comprehensive and dynamic approach to managing digital risks by:
Proactive Risk Identification: Utilizing IRM technology solutions to continuously monitor and identify emerging risks associated with generative AI. For example, detecting potential biases in AI algorithms before they can cause significant reputational damage.
Dynamic Risk Assessment: Conducting regular assessments to evaluate the potential impacts of AI and adapting risk controls accordingly. This exercise includes assessing the operational risks of integrating AI into existing systems and processes.
Cross-Functional Collaboration: Encouraging collaboration across departments to gain diverse insights and address multifaceted risks. For instance, IT and compliance teams working together to ensure AI applications meet regulatory standards.
Scenario Planning and Stress Testing: Using robust scenario planning to prepare for various outcomes and mitigate secondary risks. This might involve testing AI systems under different conditions to ensure they perform reliably and safely.
Agile Response Strategies: Developing flexible response plans that can quickly adapt to new risks as they arise. For example, having contingency plans in place to address unexpected failures in AI systems that could disrupt business operations.
The insights from the Wall Street Journal and the perspectives shared by C3.ai's CEO, Tom Siebel, are timely reminders of the importance of comprehensive risk management. By leveraging IRM technology, organizations can better navigate the complexities of digital transformation. This approach helps manage current risks and prepares organizations for future challenges, ensuring resilience and sustained performance in a rapidly changing world.
Stay tuned for more insights and strategies on managing digital risks in upcoming issues of the RiskTech Journal.
References:
Briggs, J., & Mims, C. (2024). The AI Revolution Is Already Losing Steam. Wall Street Journal.
Savitz, E. J. (2024). C3.ai Founder Tom Siebel on Why AI Is Hot and Enterprise Software Is Not. Barron's.
