With rapid developments in artificial intelligence (AI), will the job of data scientist — voted the #1 best job in America in Glassdoor’s annual report and predicted to have 2,720,000 job openings by 2020 — be obsolete in 20 years?
To answer that question, we spoke with industry experts about the career outlook for data scientists in future decades and how aspiring data scientists can guarantee their job security in this fast-changing field.
The Need for Empathy and Ethical Thinking in Data Science
The role of a data scientist is to help companies and organizations make better decisions based on the data available.
The very nature of the job requires empathy because a data scientist needs to take other people’s concerns into account when designing an analytic solution that fits their customers’ needs in a positive way.
Ryan Welsh, Founder, and CEO of Kyndi, explains how empathy and a strong ethical foundation will be even more important for future data scientists:
“I think 2018 is gonna be a big year for machine learning and ethics responsible for artificial intelligence (AI). It’s not only building the AI. Regulation is coming up for this industry in a huge way. Europe is starting to get into privacy laws and a lot of those different aspects of regulation are coming up. [Data scientists] are uniquely suited to not only have the empathy but also the skills to create good regulation.”
The Essence of Data Science Demands Good Communication
Skilled data scientists don’t just analyze data.
They also have the skills to communicate their insights in a compelling way that non-technical decision-makers can understand and act on.
Facebook Data Scientist Laura Godlewski emphasizes the critical role of data visualization and storytelling in a data scientist’s job:
“We can’t be successful without data communication; otherwise, it’s who the heck cares? You can come up with all the numbers, but nobody understands it unless someone can translate it. That is a key part of your role as a data scientist.”
Will the Rise of Machine Learning and Artificial Intelligence Automate Data Science?
In part, “Yes,” says Welsh, “A lot of the grunt work and clean-up of the data science process will become automated.”
However, automation is not necessarily a bad thing for future data scientists.
It could allow them to focus more attention on these high-level job functions:
- Asking the right questions
- Choosing the right problems to solve in their businesses and organizations
- Effectively communicating the insights derived from the data
- Understanding the ethical implications of their work
Welsh believes that some data science processes will become automated in the future. But he emphasizes that automation cannot replace the critical human competencies that make data scientists so valuable to their organizations.
“Absolutely prepare for the streamlining of these processes and data science, but it’s just cleaning up a lot of the grunt work. If you [the data scientist] can tie data to business objectives, that’s where you will [be successful].”
New Technologies Will Only Go So Far
Data scientists are tasked with solving business challenges that require high-level judgement skills, and they cannot be solved by machine learning and AI alone.
Shane McCarthy, a Data Scientist at Uber, emphasizes that automation can only take the data science process so far before the sophisticated analytic competencies of talented data scientists are used to extract the value from the data:
“You’ll still need that human touch. For a lot of our experiments and our experimental analysis, there are tools that do it for us, but that’s only [up] to a point. Can you trust this? Can you derive [the same] insights that you could if someone was actually looking at it? I don’t see there being this tidal wave of automation where the [human touch] is removed.”
In fact, as new technologies and tools emerge, data scientists trained with a sound understanding of the techniques they are using on the models they are running will be in even greater demand.
Data scientists’ deep learning empowers them to not only run code, but to develop nuanced and innovative approaches to complex data questions.
Godlewski believes there will always be a need for data scientists who are agile thinkers on the leading edge of new technologies and techniques:
“If they can automate everything, sometimes I worry, ‘Will I have a job in five years?’ But working with human behavior, we as humans change our behavior constantly. I compare hypotheses to an hourglass — you can fill your hypotheses with all sorts of great evidence until it turns on its head and suddenly you have to understand a new behavior. So, despite the automation in this field, if you stay on the front cusp of the learning and the technology, you will be in a really great position.”
How Can Aspiring Data Scientists Ensure Long Futures in their Careers?
By becoming three-dimensional data scientists.
Three-dimensional data scientists have these three key competencies that will never become obsolete:
- A strong foundation in math and statistics, and a deep understanding of the techniques they are using.
- The ability to communicate data insights to non-technical stakeholders.
- Strong critical thinking skills and a solid ethical framework that guides them in performing and managing their analytic activities.
Three-dimensional data scientists can adapt and find innovative solutions that add value to their company.
Notre Dame’s online master’s in data science prepares students to become three-dimensional data scientists by going beyond the techniques. Our program is intentionally designed to facilitate deep learning, meaning that students can go beyond the repetition of tasks, and think critically about the process.
Learn how Notre Dame’s online master’s in data science can prepare you to be a three-dimensional data scientist with a unique combination skills that will always be in high demand by employers.
Download our white paper to learn more about the 5 Trends Shaping the Rise of the Three-Dimensional Data Scientist.