Putting governance top of the AI agenda

Historically, when new ideas are created, the excitement around the potential propels the idea forward. In some cases, the euphoria becomes so all-consuming that leadership will not only greenlight the idea but push it faster out the door on concerns that the competition may be thinking about the same thing. What you have is a race to achieve first-mover advantage. Certainly, the success of companies like Amazon, Apple, Airbnb, Netflix and Tesla attest to the power of first-mover advantage.

But as technologies and processes grow in complexity and the negative repercussions of an omission become apparent, regulations have stepped in to make sure innovators think things through and cover as many of the bases as possible.

Why data governance is important

According to Gartner, as regulations evolve within the information technology industry, data governance is becoming essential for proper data management and compliance. Data governance improves data security and reduces data breaches.

Just how important is governance, especially as organisations accelerate artificial intelligence-centric innovation? Gartner warns that by 2027, 60% of organisations will fail to realise the anticipated value of their AI use cases due to incohesive ethical governance frameworks.

Acknowledging the importance of data governance in navigating a rapidly changing business environment. David Chan, managing director for AdNovum Singapore says data governance is essential for navigating complex data management challenges in a rapidly changing business environment and warns that aligning IT with business outcomes will become one of the key challenges in their governance journey.

“A clear data governance framework is therefore necessary to address concerns like data sprawl and ensuring integrity from end-to-end,” he continues. “Such governance guarantees standardised approaches, transparency, and accountability, transforming data use cases into tangible business value.

“With the recent boom in generative AI, a structured data governance program is crucial to handle the copious amounts of data being generated, contributing to analytics and machine learning.”

David Chan

Andy Ng, vice president and managing director for Asia South and Pacific region at Veritas Technologies, comments that the uptake of AI governance will drive further the importance of data, particularly in AI training.

“For example, developers must factor in how data is being selected and whether it is subject to the same principles around security, privacy and governance, similar to other business purposes. Failing to do so would expose organisations to potential data risks,” cautioned.

Why organisations struggle with data governance

Asked why enterprises struggle to implement, execute and sustain data governance processes, Ng puts the blame squarely on its complexity and difficulty to implement.

“Firstly, delineating roles for governance can be time-consuming and requires extensive documentation to keep track of responsibilities,” he starts. “Leaders also need to identify what resources they require, grappling with how new tools can enhance access to quality data, and how their data can be secured.

“Lastly, employees need to be adequately trained on how they should adhere to new governance frameworks,” he continued.

Chan attributes the difficulty of implementing and sustaining data governance to misalignment between IT and business outcomes. He cites a Gartner survey pointing to the absence of a standardised approach to data governance as further complicates matters.

“55% of decision-makers say they struggle with translating data use cases into tangible business value. Transparency and accountability issues, such as unclear data collection processes and ownership, also further complicate effective governance,” he posits.

Chan suggests that to overcome this, addressing data ownership, defining roles, and outlining processes for collection, curation, and qualification are crucial towards unlocking data's full potential in business operations.

Data governance boss

Asked who should be in charge of data governance in the organisation, Ng says that effective data governance is ultimately a team responsibility – and that includes business executives, data management professionals, IT staff and end users who are familiar with relevant data domains across systems.

“The Chief Data Officer sets the strategy and oversees the framework, whereas the data governance teams manage specific data sets, ensuring quality and adherence to policies. Business leaders are responsible for enforcing policies, while all employees play a key role by adopting the best practices based on established guidelines.”

Andy Ng

Concurring with this view, Chan says effective data governance requires collaboration among various stakeholders in an organisation. C-Suite leaders provide strategic vision and allocate resources for a robust framework, emphasising a data-driven culture. Data Governance or IT teams develop and implement policies, ensuring consistency and quality. Employees adhere to governance policies, provide feedback, and ensure ethical data use.

“Each stakeholder plays a critical role at different stages, necessitating a collective effort for successful implementation, execution, and long-term maintenance of data governance practices within the organisation,” he elaborated.

Achieving a sustained data governance practice

Forrester says businesses must constantly adapt to cyclical post-pandemic macroeconomic conditions, emerging technologies like generative AI, and new consumer trends like social commerce. For leaders to deliver the next wave of business growth, they will need superior and relentless alignment to achieve high-performance IT — the pursuit of continuously improving business results through technology.

Veritas’ Ng says lasting data governance hinges on cultural change, not just policy implementation. “Fostering a data-driven mindset across all levels, where individuals understand the importance of responsible data use and hold themselves accountable, is crucial for building a sustainable framework,” he posits.

He concludes that this cultural shift empowers employees to become active participants, not passive players, in data governance efforts.

Adnovum’s Chan believes that for an enterprise to achieve lasting data governance, it is crucial to establish a transparent and accountable data governance framework tailored to the organisation's specific business needs and objectives on their use of data.

He reasoned that by contextualising data challenges within the broader business goals and outcomes, organisations can determine the relevance of data in everyday activities.

“Additionally, ensuring data integrity, implementing data quality metrics, and adhering to regulations are essential components, to not only safeguard the data but also enable the organisation to unlock the full potential of insights and business value across diverse use cases, includes Chan.