Diving into the AI talent pool.
ChatGPT promises to democratize artificial intelligence, and is already making it relatively easy for non-data-scientist types to partake in the wonders of machine-generated wisdom. Google and Microsoft are following suit and souping up their search engines. Does this portend reduced demand for AI talent?
Relief from AI talent shortages isn’t likely anytime soon, but democratized AI may expand the meaning of business intelligence. “We are trying to teach business users to speak AI instead of teaching AI to speak business,” says Arijit Sengupta, CEO and Founder of Aible. “Before sending business users to learn Python, stop and say: ‘why can’t the AI understand my business needs and generate the Python code automatically?’ The Internet revolution didn’t happen because everyone learned how to code to interact with the World Wide Web; it happened because the Netscape browser could be used by almost anyone.”
The more AI, the hotter the demand for people to build and refresh it. Organizations are struggling to make it work, the latest McKinsey research shows. “We might be seeing the reality sinking in about the level of organizational change it takes to successfully embed this technology,” says Michael Chui, lead author of the survey report. In 2017, 20% of respondents reported adopting AI in at least one business area. After peaking at 58% in 2019, it’s dropped to 50% today.
Leaders across the industry agree that more talent is needed to move forward with AI. “Talent is a serious barrier,” observes Dr. Vishal Sikka, founder and CEO of Vianai Systems. “There may only be about 20,000 to 30,000 people in the world that understand the true methods of how AI systems run. This is vastly smaller than the 52,000 or so people we estimate are MLOps professionals, or the 1 million we estimate are data scientists. Many of them could not tell you why the system is doing what it is, why it makes the recommendations it does, what could possibly run awry, or how the underlying techniques work.”
AI talent shortages are being felt acutely within the financial services sector among others. Recruiting and retaining data scientists is identified within a survey from NVIDIA as the top obstacle to AI in financial services — 36% of executives report such difficulties — an increase of 80 percent over last year.
“There is a vast asymmetry today between organizations actually investing or wanting to invest in AI, and the talent available that understands the technology,” Sikka says. “This lack of available talent to explain AI and provide the necessary oversight, leads to issues of trust, bias, and transparency — key concerns for businesses that want to implement AI. They need to ensure bias isn’t introduced into the AI models. The key to an equitable system is the assurance there is intelligence built into the system to ensure that it is fair, trustworthy, and that the model hasn’t started to drift or build upon biases such as race, gender, socioeconomic status, and other factors.”
Companies are struggling “to find suitable candidates that know enough about AI,” agrees Shalabh Singhal, CEO of Trademo. “They also suffer from hiring the wrong talent; someone might be skilled in AI but not suitable for their use case or the stage of the company in its AI journey. This can hurt the lifecycle of AI projects and lead to operational difficulties.”
A typical AI project requires a highly-skilled team “including a data scientist, data engineer, machine-learning engineer, product manager, and designer—and there simply aren’t enough skilled professionals available, even with the recent contraction across the technology industry,” the McKinsey authors state. “When it comes to sourcing AI talent, the most popular strategy among all respondents is reskilling existing employees. Nearly half of the companies we surveyed are doing so.”
Companies of all sizes “need to increase their tech literacy enterprise-wide to create a wider range of talent working on the AI systems in place,” Sikka says. “More employees must be educated on the transcendent aspects of AI in particular. They need to learn the limitations and the weaknesses. Not just what it can do, but things that it cannot do and what needs to be built in an AI system that compensates for these limitations.”
And yes, generative AI such as ChatGPT is helping to smooth the way for preparing more people for an AI-driven future. “The newest wave of generative AI models promises to reinvent functions such as communications, sales, and human resources,” the McKinsey authors relate. “As individual AI capabilities, such as natural-language processing and generation, continue to improve and democratize, we’re excited to see a wave of new applications emerge and more companies capture value from AI at scale.”
The successful implementation of AI “requires constant improvement enabled by the robust feedback mechanism,” says Singhai. “Patience is key while awaiting the business benefit from AI. Business leaders must invest in the people working on AI projects to ensure they understand the ultimate goal. With an upskilled workforce and opportunities for improvement, while keeping a close eye on the pitfalls, businesses reap the rewards of their AI investments.”