Generative AI: A Game-Changer for Tech Leaders

Every day, the media is abuzz with ground-breaking developments in the realm of generative AI, revolutionising the business landscape. This fervour is entirely justified; according to McKinsey’s research, Generative AI (GAI) has the potential to contribute an annual value ranging from $2.6 trillion to $4.4 trillion.

After engaging in extensive dialogues with numerous technology leaders and scrutinising over 50 generative AI projects across various companies, including our own, McKinsey identified a comprehensive set of nine strategic actions that technology leaders should embrace. These actions are designed to foster value creation, orchestrate the utilization of technology and data, facilitate scalable solutions, and adeptly manage the associated risks of generative AI.

Define the Company’s Stance on Generative AI Adoption

As the use of generative AI becomes increasingly prevalent, some CIOs and CTOs have chosen to restrict employee access to publicly available applications as a risk-mitigation measure. However, such caution may inadvertently stifle innovation and hinder skill development among employees.

Identify Use Cases for Value Enhancement

CIOs and CTOs should steer their organizations away from the overwhelming pursuit of numerous use cases without clear benefits. Instead, they should collaborate with CEOs, CFOs, and other business leaders to explore how generative AI can challenge existing business models, unlock new opportunities, and generate fresh sources of value. For example, McKinsey’s research indicates that generative AI can boost productivity in certain marketing tasks by approximately 10% (analysing unstructured customer preference data) and customer support by up to 40% (e.g., using intelligent bots).

Reimagine the Technology Function

Generative AI has the potential to revolutionise how technology functions and operates. CIOs and CTOs should thoroughly assess the potential impact of generative AI across all technology domains. Action should be taken swiftly to build experience and expertise in three primary areas:

1. Software development: McKinsey’s research demonstrates that generative AI coding support can accelerate software engineers’ coding speed by 35-45%, code refactoring by 20-30%, and code documentation by 45-50%.
2. Technical debt: Addressing technical debt, which can consume 20-40% of technology budgets and slow development, should be a priority.
3. IT operations (ITOps): CIOs and CTOs should examine how generative AI can expedite ITOps processes.

Leverage Existing Services or Adapt Open-Source Generative AI Models

Choosing whether to develop generative AI capabilities in-house or rely on existing services resembles the classic “rent, buy, or build” decision. The guiding principle is to invest in proprietary generative AI capabilities where they confer a competitive advantage and utilize existing services for more standardized tasks.

Enhance Your Enterprise Technology Architecture to Accommodate Generative AI Models

Organisations will deploy various generative AI models differing in size, complexity, and capability. To derive value, these models must seamlessly integrate with existing systems and applications. Consequently, establishing a separate tech stack for generative AI introduces unnecessary complexities.

Establish a Data Architecture for Quality Data Accessibility

The ability to harness generative AI models for generating value, including cost savings and improved data and knowledge protection, hinges on effectively utilising an organisation’s data resources.

Create a Centralised, Cross-Functional Generative AI Platform Team

In line with the trend toward product and platform operating models, CIOs and CTOs must integrate generative AI capabilities within this framework. A crucial initial step involves establishing a dedicated generative AI platform team tasked with developing and maintaining a platform service for provisioning approved generative AI models on-demand for product and application teams.

Tailor Upskilling Programs to Role and Proficiency Levels

Generative AI has the potential to significantly enhance employee productivity and capabilities, but the benefits vary based on roles and skill levels. Thus, leaders must redesign skill-building initiatives to align with specific job requirements. Our recent empirical research, employing the generative AI tool GitHub Copilot, demonstrated that software engineers could write code 35-45% faster.

Evaluate the Evolving Risk Landscape and Implement Ongoing Mitigation Measures

Generative AI introduces a fresh set of ethical and operational risks, such as “hallucinations” in AI responses, accidental release of confidential personal data, inherent biases in the datasets used, and IP-related uncertainties. Addressing these concerns requires vigilant evaluation and the establishment of continuous mitigation practices.

GAI is poised to become one of the most rapidly evolving technology domains. Consequently, tech leaders cannot afford to delay defining and shaping their generative AI strategy. While the landscape continues to evolve at a breakneck pace, embracing these nine actions can empower CIOs and CTOs to responsibly and effectively harness the immense potential of generative AI on a grand scale.