By Mudit Srivastava
It is common knowledge that AI will create unprecedented innovations, and most people realize it as we see its examples in our day-to-day lives. But understanding industry needs and expectations from the talent of tomorrow is nebulous at best. It is becoming increasingly evident that AI will be used to augment rather than replace human intelligence. A key component of succeeding in an AI-powered world is knowing how and where to harness its power. That is where AI education comes in.
If we look at the industry, AI adoption is not close to where it will be a decade from now. AI has the potential to create trillions in economic wealth and solve some of the biggest challenges of humanity, including epidemics, education, health, and much more. But talent availability is still a challenge. Ninety percent of data science recruiters struggle to find good data science talent due to a massive shortage. Where is this gap?
A decade ago, companies were beginning to analyze voluminous and less-structured data such as clickstreams, social media, and images and speech. The most common qualification in a popular HBR survey of that time included PhDs in some scientific field, exceptional grip on math, and knowing how to code. The goal was to hire talent who could wade through complex, messy datasets and build recommendation engines.
Today the job of a data scientist is way more in demand, but its nature has drastically changed. This is because the technologies data scientists use are making huge strides, and organizations are in the process of learning the best data practices. The use of open source tools is increasing, and some advancements affect the core of data science work, from data preparation to AI deployment. Furthermore, the importance of non-technical expertise, such as ethics and change management, is growing.
In such changing scenarios, the industry needs to truly democratize AI education and bring more people who can quickly learn skills and implement them to deploy models. These skills do not just include technical know-how and go beyond it. For example, a vital aspect of the deployment of any AI is managing change in the business process. A successfully implemented AI could transform the set of business or technology-related tasks that had previously been manual, heuristics-based, or simply impossible. This necessitates skills such as stakeholder analysis and change management, which are comparatively undercovered in traditional data science training programs, leaving many data scientists unprepared. Some organizations have the liberty to leave data scientists to focus on creating models and coding while leaving organizational or technical deployment issues to other roles. But whether that’s the approach that will stay when AI adoption reaches its peak cannot be said.
On LinkedIn, there are more than 300K data scientist job openings, and roughly half of them are located in the United States. As per a U.S. Bureau of Labor Statistics prediction, data science will grow more than most fields by 2029. This shows the opportunity for other emerging talent markets to capitalize on the unaddressed industry needs.
In a world characterized by widespread AI adoption, everyone will need a primary AI education to compete and succeed. The good news for people who want to transition full-time into data science is that their need is very much, even today. And it is bound to continue that companies will continue to seek and build diverse data science teams, which will create room for people from different functions to enter into AI/ML. However, the AI learning frameworks need to be agile, and talent needs to focus on building the right skillsets rather than relying entirely on existing academic or professional educational institutions.
Companies want to use advanced analytics and AI within their organizations profitably. However, a shortage of industry-ready young talent and leadership is impacting its pace. AI is the key to solving some of the biggest challenges with AI, which are now bottlenecked due to a lack of quality talent and awareness of best data practices. If AI education needs to reach the right set of people, money, of all things, should be the last bottleneck because the opportunity cost due to lack of talent will be very high. And talent and enterprises should be aware that if they want to train in data science capabilities, there will be plenty of options. It is unlikely that any single course or boot camp will inculcate all of the skills necessary to conceive, build, and deploy practical and ethical AI analysis, experiments, and models. There needs to be a continuous process where individuals and institutions take proactive measures to upskill.
The author of this article is founding team member, Dphi. Views expressed are personal.