Growing out of control?
If 2023 was the year of AI exploration, then 2024 will be the year of action. According to Forrester, investment in AI software will grow 50% faster than the wider software market. A recent PagerDuty survey also found that 71% of businesses are looking to expand investments in AI and machine learning (ML) in the next year. This rapid investment in, and expansion of, AI tools within businesses poses a particular headache to CIOs, who must ensure that everyone in the organization uses AI in a compliant way.
Given that training data is the foundation for all GenAI models, organizations must ensure the cleanliness and trustworthiness of their data, and that management of data is ethical. With improving security, reducing risk, and driving revenue growth among organizations’ priorities for the use of AI, there are a number of factors CIOs will have to consider when outlining AI strategy.
A question of data
Data is the core building block of AI. The data fed into an AI model in its training phase is the only source of information it is able to draw from throughout the model’s lifecycle and dictates the model’s usefulness. When creating these datasets, it is vitally important for CIOs to ensure that the data their organization leverages makes sense ethically and morally. At the same time, they must also be able to make an effective business case for leveraging this data. This will help to ensure that their organization uses AI in a compliant manner, which can be especially challenging with the breakneck pace of changes in AI regulation.
To support with this compliance process, CIOs must implement effective data management processes throughout their organization. Traditional, siloed approaches must be amended to allow data to flow between teams within each organization. Cross-departmental information sharing needs to become the norm for effective AI usage. After all, when a GenAI model is trained on the vast amounts of data and historical insights that exist across an entire organization, it can provide intelligent recommendations and automate repetitive tasks. This can hugely reduce the cognitive load of time-intensive processes, such as incident resolution.
Risk and opportunity: A crucial balancing act
The most common top priority for technical leaders is improving their organization’s security (29%). With this goal in mind, CIOs must take care to ensure that the introduction of new GenAI tools does not put their business in jeopardy.
Before any teams begin to use GenAI tools, there must be a thorough risk assessment of how these will affect operations. Given the amount of sensitive data that can be fed into GenAI tools, this requires an appraisal to identify and mitigate potential risks across each department within an organization. The factors that this assessment must cover include data security, privacy, fairness, and accountability. Above all, though, the key consideration throughout this risk assessment must be that the use of GenAI should never compromise your organization’s ability to do business: it must be an enabler of business success.
What does a good GenAI strategy look like?
With 72% of organizations prioritizing investment in AI and ML, it is vital for CIOs – regardless of their industry – to look to AI trailblazers for the building blocks of their GenAI strategy – as well as to understand where they should leverage AI, and where they shouldn’t.
There are three steps CIOs must take to lead the development of their organization’s GenAI strategy and ensure that any GenAI tools that are used bring a direct benefit to the business. The first is to develop a strong framework for data management and security, forming a cross-functional group that brings together members of different teams. This must include the organization’s compliance, risk, legal, security, and data teams, but can also be extended to include marketing, and other departments that may use GenAI to support their work.
The second step is to identify business opportunities where GenAI can be a win for the company or for customers so that the use of any GenAI tool is supported by a strong business case, rather than using AI tools for their own sake. Finally, CIOs must be willing to experiment and explore ‘the art of the possible.’ While this should happen alongside previously defined GenAI use cases, it can also offer an opportunity for innovation using GenAI that can drive huge benefits for an organization.
To learn more, visit us here.
