Future Says is Back for Season 3!

Despite AI dominating the headlines for almost a decade now, it continues to be a poorly defined and understood technology. Future Says was my way of getting answers. I invited some of the industry’s brightest minds to join me for informal discussions about the ins and outs of AI – what should we be excited about and what should we be concerned about?

Each episode has brought its own uniqueness. In Season 1, we learned how Telefonica is using data for social good, why DNB Bank is building an ESG Data Task Force, what a data mesh entails at Scania, the data-driven business models being developed at Husqvarna, and how H&M strive to keep people in the loop. Moving forward, in Season 2, leaders from Google, Capgemini, PWC, Fasanara Capital, Altair, Zenseact, King, and the NHS spoke about fostering diversity and inclusion, building responsible AI frameworks, augmenting cryptocurrency decision-making, utilizing gaming as an AI proving ground, the continuous embrace of the cloud, dealing with the healthcare industry’s data privacy challenges, and much more.

The Data Science Field: Challenges and Opportunities

But while my guests spoke about a variety of topics, they often referred to a common challenge – finding experienced, specialized data scientists that can handle their rapidly growing data science needs. Recent data backs this up; according to Glassdoor, data scientists are the most sought-after employees in the workforce today. Moreover, according to the World Data Science Initiative, the number of jobs requiring data science skills is expected to grow by 27.9% by 2026, and data engineering job postings have grown by an astonishing 88.3% within the past few years. But supply is falling far short of demand. In the U.S. alone, Deloitte estimates there’s a shortage of 250,000 data scientists, and the British government released similar findings in the National AI Strategy Report released last year.

However, as data science technology and tools evolve, there’s reason to believe that more citizen data scientists – people with basic data literacy skills and advanced domain knowledge – will emerge and fill these much-needed roles. That’s largely because of the rapid growth of low- and no-code coding software tools.

Gartner has found that by 2024, 65% of enterprise software development will be using low-code platforms

Gartner also predicts that the market for citizen data scientist roles will grow five times faster than that of the traditional data scientist market. My guests have backed up these statistics with examples. In Season 1, Piab’s Girish Agarwal spoke about how “there are 22,000 mechanical engineers at Husqvarna that can configure a piston but not an AI model”. Girish said that if the company could make all these engineers more data literate, they could achieve outstanding innovation alongside a smaller team of collaborative data scientists.

When we think about how technology-related entry barriers fall, the rapid rise of low- and no-code software tools shouldn’t come as a surprise. We’ve seen a similar story so many times. In my day-to-day, I am constantly amazed at the democratisation of simulation technology throughout engineering. Finite Element Method is essentially the “code” and principle which forms the backbone of almost all modern simulation solvers today. This started out like data analytics – code and research in universities – but there are now lots of different GUI’s for this work. Today, almost all engineers are expected to perform some level of simulation at every stage of the design lifecycle. I expect the same to be true of data analytics in the near future.

Building a Smarter, Data-Driven Future

Clearly, the key to the future of AI is building teams of people that have collaborative, overlapping skillsets. In Season 1, Errol Koolmeister, former head of AI engineering at H&M, discussed how the organization aimed to build teams of people with “T-shaped skills profiles.” In other words, this means that people should have a deep understanding of one field, but also have a basic understanding of many others as well. Furthermore, Telefonica’s Dr. Richard Benjamins emphasised the need for teams that satisfy:

“four C’s: creative thinking, critical thinking, collaboration, and communication.”

And in Season 2, many guests talked about ways they’re training their employees to ensure they’re upskilling in the best ways possible. Capgemini’s Niraj Parihar discussed how the organization has started an AI academy using ranks to indicate progress, such as “cadet, genie, guru, and captain.” The NHS has also invested in its employees’ skills, as Ming Tang described how she’s given her team regular free time for “self-directed learning”.

Season 3

Of course, finding people with the right skills and building well-rounded teams doesn’t come easily. That’s why Season 3 of Future Says will be laser-focused on this. All of this season’s guests have started in engineering and made career transitions towards AI. They have all been on this journey to combine domain expertise with data understanding:

• Vijayakumar Kempuraj, Digital Twin Lead, Ford Motor Company
• Ravi Parmeswar, Vice President of Business Intelligence, Johnson & Johnson
• Jan Chirkowski, Vice President of Analytics and Fleet Operations, Kongsberg
• Fran?ois Deheeger, Senior Fellow of AI and Data Science, Michelin
• Mieke De Ketelaere, Adjunct Professor of Sustainable, Ethical, and Trustworthy AI, Vlerick Business School and Strategic Advisor to IMEC
• Jada Smith, Global Engineering Director of Software Platforms, Aptiv

In previous series, we have learned about industries like banking, gaming, space, retail, and healthcare. As much as there were similarities, each industry is facing their own challenges, so I wanted to break out one in particular – that of engineering and manufacturing. For me, this is the industry that is lagging in digital maturity. According to a recent report, manufacturers still use paper for 35% of their processes, and nearly half (49%) use manual spreadsheets. As such, Season 3 will emphasize how to bring processes and products into the digital age, how to build good data infrastructure, how to modernize operations and organizations, and more. Last season, we heard from Zenseact’s Vanessa Eriksson about how autonomous cars are now generating up to 50 terabytes of data per car, per day! If one car can generate that, think about all the insights we could get from the entire product lifecycle!

What started as a personal journey has become something much bigger and I hope you are as excited as I am to continue this conversation on what the future holds for AI!

Sign up here to get the latest updates and watch any episode you like on-demand.