Artificial Intelligence (AI) is a part of our lives, whether we like it or not, whether we think we use it or not. Somewhere along the chain of production in our daily activities – shopping, working, searching online – we encounter AI in one form or another.
However, as AI becomes an increasingly integral aspect of business operations, so too must appropriate governance. Businesses must endeavour to understand issues and challenges to AI adoption so that operational gaps do not form. In understanding its full potential, businesses will have greater control over AI application in industry. After all, better the devil you know.
AI is here to stay
To survive in today’s climate, companies must ensure transformational capacity and ability is in-built. Digital transformation may have become something of a buzzword, but for good reason. It is seen and heard everywhere because it is important to all facets of the modern enterprise. To be an industry leader, businesses need to demonstrate the capacity to adapt their processes and alter business models in line with cultural and technological shifts.
AI and machine learning represent such a shift. The technology has seen a usage boom in recent years. Notably, a majority of organisations are evaluating AI or using it in production. In fact, over half of respondents in our recent survey on AI adoption in the enterprise identified as “mature” users of AI technologies – that is, they’re using AI for analysis and/or in production. Only 15 per cent of respondents reported that they’re not using AI at all.
Unsurprisingly, across the board research and development dominate in current AI adoption trends, followed closely by applications in IT and customer service. That being said, respondents cited a widening range of industry areas in which functional parts of a company use AI. As a whole, this indicates that companies are increasingly turning to AI and machine learning as a business tool.
Change is challenging
Obstacles are to be expected on the path to digital transformation, particularly with unfamiliar entities in the mix. For AI adoption, the most prevalent obstructions are: a company culture that doesn’t recognise a need for AI, difficulties in identifying business use cases, a skills gap or difficulty hiring and retaining staff and a lack of data or data quality issues.
With such a broad spectrum of challenges, it is worth delving into a couple of them. Firstly, it is interesting to note that an incompatible company culture mostly effects those companies that are in the evaluation stage with AI. When rephrased, perhaps it is obvious – a company with “mature” AI practices is 50 per cent less likely to see no use for AI. By contrast, in a company where AI is not yet an integrated business function, resistance is more likely. Secondly, AI adopters are more likely to encounter data quality issues; by virtue of working closely with data and requiring good data practice, they are more likely to notice when errors and inconsistencies arise. Conversely, companies in the evaluating stages of AI adoption may not be aware of the extent of any data issues.
Yet, despite data quality issues being so frequently cited and clear patterns in when they typically occur, the very fact that year-on-year it remains a top concern indicates that many companies do not consider data governance a priority.
Good governance means good data
Of the nearly 1,400 respondents in our survey, just over one fifth stated that they have implemented governance procedures and/or tools to support and complement their AI projects. It is evident, then, that many businesses are yet to initiate measures to control common risk factors and embed formal processes for data governance and conditioning – a task made more arduous the longer it is postponed.
In this case, then, AI “immaturity” may well work in a business’s favour. The benefit of nearly half of companies identifying as being at the early stages of AI and ML adoption is that the introduction or implementation of proper governance and risk management may be less of a culture shock and, therefore, should more easily be adopted. More good news is that over a quarter of companies reported that they have plans to introduce formal data governance by 2021.
However, the question still remains – why aren’t companies building data governance into their AI projects from the outset? Data governance, like any type of business governance, should be embedded within corporate strategy. Particularly in recent years, with many companies now adopting digital business models, data management should be optimised and championed.
Data is the driver for AI and digital transformation. Yet time and time again, we see instances where it is not leveraged in a way that reflects its value. Of course, it is never as easy as we want – data governance and conditioning take time and resources. However, it must be viewed in terms of the benefits it will bring: observability, reproducibility, efficiency and transparency.
Don’t let governance be an afterthought
With AI now very much a part of business function and only set to increase in reach and take-up, the enterprise must react accordingly. In understanding the main challenges and obstacles to AI adoption, companies can proactively look to tackle them. Moreover, for companies yet to begin their AI journey, prior knowledge of challenges will allow them to prepare and plan. Addressing company culture and practices early on makes a big difference down the line. Many have had to learn the hard way, so businesses should take heed when they can.
It is essential that data governance procedures are given the careful consideration they require and that – as much as possible – companies avoid viewing them as an addendum tacked on to digital transformation plans. As is the case with other governance structures, data governance requires time and resources.
Data is at the heart of AI and digital transformation and, moving forward, will be the key to success and scalability. We talk about AI integration constantly but now must also talk about data governance integration.
Rachel Roumeliotis, Vice President of Data and AI, O’Reilly