As we look ahead to the next 12 months, we can see a lot of trends continuing from 2023 when it comes to data management and implementation. We’ve seen AI take hold of boardrooms and social media feeds alike, Data & Analytics teams reduced as they fail to demonstrate measurable value, and a clear need for regulations to cover the good and potential nefarious uses of AI technologies.
Clean data is a professional advantage
If we’ve learned anything from 2023 it’s that data separates the wheat from the chaff, and those taking a flippant view will be left in the dustbowl. As Susan Walsh, The Classification Guru, perfectly phrases it, “If [your model is] pulling from dirty data, you’re going to get dirty answers.”
And clean data is the enabler of AI in business, with use cases across several functions including marketing, customer satisfaction, procurement and more. Using data in this way can leverage opportunities like never before at breakneck speeds, likely reducing costs. Understanding competitors and customer needs are not to be ignored. But this can only be harnessed productively if the underlying data you’re working with is reliable, performant, and available.
Perhaps the most important point is that data is the irreducible foundation of AI, analytics, and value. To get your data management wrong is to scupper your organisation’s chances in the year ahead. Get it right and you pave the way for improved capability and decisioning.
If [your model is] pulling from dirty data, you’re going to get dirty answers
Data Analysts at risk?
The economic uncertainty of 2023 is likely to carry over to 2024 with organisations looking to streamline costs. And as we move to data being packaged up as a product or service, organisations will need to determine how profitable the data products and the teams responsible for them are.
Kyle Winterbottom, CEO of Orbition, states “in any economic uncertain times, anything that is deemed to be a cost centre that cannot articulate the value that it’s adding is going to be at risk of being cut.”
So, these teams need to prove their value and articulate it in ways that make sense to CEOs, or face being cut as businesses keep up with the current economy. Which can seem counterintuitive as we talk about how crucial analytics are, but when AI can process massive data sets in a fraction of the time, automation could mean some roles are changed with the evolving landscape.
But using AI in this way opens up a whole new concern, with Stijn Christiaens, Co-Founder of Collibra, pointing out that with each entry point and decision made comes the opportunity for a mistake or a security breach. Data Governance will remain a top priority for teams in 2024.
Data Governance will remain a top priority for Data & Analytics teams in 2024
AI governance and safety
Of course, you need to be able to leverage AI safely, securely, and confidently. We predict that AI governance and regulations will come into action across the globe, following the US’s Blueprint for an AI Bill of Rights and the EU’s very horizontal regulation The EU AI Act. Stijn Christiaens predicts that an AI Act in Europe will be “forcing organisations to do better with AI and data,” and that this likely “comes with fines as big as the GDPR days.”
These regulations need to be created with the input of data scientists around the globe, who understand the sheer power that has been made so readily available, yet is still so casually used in the mainstream.
And when mistakes are made, however grave the consequences, we must be certain with whom the blame and consequences lie. Of course, the machine itself cannot be to blame – the operator or manufacturer, however, can be. We don’t, for example, blame the chatbot for being racist or sexist, as Stijn Christiaens posits, but those failing to spot (or wilfully ignoring) the damaging bias present in the training data.
This is something we need to be teaching our children as well as our CEOs
Data Fluency training across the board
“This is something we need to be teaching our children as well as our CEOs.” – Stijn Christiaens
As we’ve touched on in this article, AI models process information at face value. What’s true is true and is not questioned. Therefore, the operators and manufacturers are responsible for challenging the outputs, thinking critically and pragmatically in a way that (as of now) only humans can do.
This directly translates to the need for advanced data fluency and literacy at every level, along with invaluable critical thinking skills. Innocent mistakes could, after all, be just as damaging as nefarious ones.
Education is key as data science spreads through the echelons of an organisation to those who may have never dealt with the intricacies before, and those at all levels of an organisation find themselves in the same boat. This education is likely to be personalised to each role and help everyone within an organisation, and indeed society, understand the role and importance of data safety.
We predict that this will become an essential part of 2024 – the collaboration between CDOs & DA teams, and the wider business as it becomes more apparent that you’re all on the same team, and a more holistic approach is needed to progress.
Don't forget the foundations
Whatever 2024 brings, the need for clean, well-governed data hasn’t gone away. Whether you’re getting the basics in place or wanting to take advantage of all the latest advancements, help your organisation thrive by getting your data right.
contentS
- Clean data is a professional advantage
- Data Analysts at risk?
- AI governance and safety
- Data Fluency training across the board
- Don't forget the foundations
Webinar
We discussed this topic – and much more! – with our fabulous guest speakers in the 6th online event of our Data Leadership Series. Catch up now!
