Application modernization is a top priority for many organizations, and generative AI has the potential to transform this process by automating tasks such as code reverse engineering, code generation, and code conversion. However, before adopting generative AI, CIOs and CTOs need to carefully evaluate their technology and data estates, assess the costs and benefits, and develop a strategic plan. This article provides a guide for CIOs and CTOs on what to consider before adopting generative AI for application modernization.
Key Considerations for CIOs and CTOs
Before implementing generative AI for application modernization, CIOs and CTOs should consider the following:
- Evaluate their organization’s level of hybrid cloud mastery as a foundational strategy for implementing generative AI effectively.
- Assess the organizational obstacles and costs of implementation and maintaining the status quo.
- Weigh the costs and benefits of using general-purpose large models versus tuning smaller ones.
- Assess factors and costs related to data availability, governance, security, and sustainability.
- Collaborate with HR to ensure that people are at the center of the generative AI strategy.
Hybrid Cloud Accelerates Generative AI Adoption
Implementing generative AI on a hybrid cloud infrastructure can provide organizations with the agility and scalability needed for application modernization. Hybrid cloud allows organizations to use open-source large language models, leverage public data and computing resources, and fine-tune models securely while keeping proprietary insights private. This approach not only enhances customer and employee experiences but also enables IT automation and continuous modernization.
The Business Context
Even for organizations committed to hybrid cloud, there are obstacles to application modernization that need to be addressed. CIOs and CTOs should estimate the financial impact of modernization across the organization and champion it as a business initiative. They should also prioritize talent development and cultural buy-in to ensure that modernization is seen as a strategic investment rather than just an IT project.
Prioritizing Generative AI Use Cases
Generative AI can be applied to various IT operations, including automatic triaging of systems, managing queries and tickets, and event detection and management. It can also assist in transformation planning, code reverse engineering, code generation, and code conversion. CIOs and CTOs should focus on these use cases and start with proof-of-concept implementations to explore quick wins and test scalability.
Evaluating Foundation Models
Choosing the right foundation model is crucial for accurate and efficient outcomes. While larger models often produce better results, a smaller model that has been fine-tuned for a specific task can outperform a larger model. CTOs should consider the natural and programming languages supported by the model and the amount of fine-tuning required.
Creating a Customized ROI Framework
Calculating the return on investment (ROI) for generative AI can be challenging due to the lack of mature methods and comparative benchmarks. CIOs and CTOs should consider pricing and licensing methods, development effort, enterprise data security, and potential IP and security risks when selecting and deploying a model. They should also assess the impact of data availability and governance on ROI.
Consider Sustainability Goals
Sustainability is an important consideration in generative AI adoption. Training and running AI models can have a significant carbon footprint, so CTOs should prioritize energy efficiency and develop strategies to reduce expense and carbon emissions. IBM has developed energy-efficient model training technologies to help address this issue.
Lead with Human Resources
The successful implementation of generative AI relies on skilled and enthusiastic people. HR departments should be involved in the strategy and engage in reskilling HR professionals. Creating a management initiative to communicate generative AI progress and gather feedback is also essential.
By considering these factors, CIOs and CTOs can make informed decisions about adopting generative AI for application modernization, leading to improved efficiency, productivity, and business value.
- IBM Research: What is generative AI?
- IBM Community: Harnessing generative AI for modernization
- IBM Research: Generative AI for enterprise
1. What is generative AI?
Generative AI refers to the use of artificial intelligence techniques to generate new content, such as images, text, or code, based on existing data or patterns. It can be applied to various tasks, including application modernization, where it can automate processes such as code reverse engineering, code generation, and code conversion.
2. Why is hybrid cloud important for generative AI adoption?
Hybrid cloud enables organizations to take advantage of open-source large language models, leverage public data and computing resources, and securely fine-tune models while keeping proprietary insights private. This infrastructure provides the scalability and agility required for successful generative AI adoption in areas such as application modernization.
3. How can generative AI be used in application modernization?
Generative AI can be applied to various aspects of application modernization, including transformation planning, code reverse engineering, code generation, and code conversion. It can automate processes such as summarization, plan creation, generating reference architecture, analyzing code for business rules and refactoring opportunities, generating code snippets, and converting code between different programming languages.
4. How can CIOs and CTOs evaluate the ROI of generative AI?
Calculating the return on investment (ROI) for generative AI can be challenging due to the lack of mature methods and comparative benchmarks. CIOs and CTOs should consider factors such as pricing and licensing methods, development effort, enterprise data security, and potential IP and security risks in their ROI framework. They should also assess the impact of data availability and governance on ROI.
5. How can sustainability goals be incorporated into generative AI adoption?
Sustainability is an important consideration in generative AI adoption. CTOs should prioritize energy-efficient model training and resource utilization to reduce the carbon footprint of AI models. IBM, for example, has developed technologies to recycle small models into larger ones, significantly reducing training time, cost, and carbon emissions.