No One Can Predict AI’s Impact On Jobs; But Here Are 4 Possibilities

Where are we going with AI?

getty

Does anyone really know what artificial intelligence will do to jobs and the workplace four years from now, four months from now, or even four weeks?

We hear a lot about the direction jobs and workplaces will go over the next few years, with AI leaders such as Anthropic’s Dario Amodei predicting a super-powered AI poised to subsume everything, even what were considered “safe” occupations such as nursing, to more optimistic readings such as a Yale University analysis which says there is no evidence of any type of job apocalypse on the horizon any time soon.

There is even a well-reasoned argument that all predictions so far have been off base, and no one really has the data to know what’s around the corner. That observation, courtesy of Thomas Davenport and Miguel Paredes, seems to make the most sense amid all the noise and panic.

The best we can do for now is to paint likely scenarios that may unfold, with the possibility that there may be dark spots as well as bright spots in the coming years. With that in mind, researchers for the World Economic Forum published a summary of four scenarios that they are seeing around the corner, which in this case, is the year 2030. The WEF team tapped into their network of chief strategy officers to offer the following possibilities:

Scenario 1: Extremely Optimistic – Supercharged Progress with Rapid Adaptation

“Many jobs have disappeared, but new occupations emerge and scale fast, in part with humans directing portfolios of capable machines and becoming agent orchestrators,” the report states. In the process, industries, business
models and workflows are transformed, with rising productivity and
innovation.

Scenario 2: Somewhat Chaotic – With Uneven Adaptation

AI advances exponentially, but members of the workforce have difficulty keeping up. “Businesses race to automate as a stopgap, displacing workers faster than education and reskilling systems can respond,” the researchers predict. Agentic AI dominates, but risks associated with unemployment and eroding confidence grow.

Scenario 3: People and Machines Learn to Get Along, Productively

The WEF researchers call it a “co-pilot economy,” marked by gradual AI progress that augments current workforce capabilities. “The AI hype of the 2020s has
given way to pragmatic integration” – at last. “Most industries see incremental transformation as human–AI teams reshape value chains.” Keys to success include investments in training, mobility, digital infrastructure and AI governance.

Scenario 4: Frustration and Dysfunction

Any progress associated with the above three scenarios stalls out, hampered by lack of critical workforce skills due to undertraining, and disappointing productivity. “Businesses lean on automation to backfill scarce talent. Displacement hits primarily routine roles, while the value of skilled trades and manual occupations increases. The hope of AI-enabled prosperity fades into frustration, as adoption gaps fuel inequality, create a bifurcated economy and limit growth.”

Not considered in this report is the impact of a global economic downturn, which would likely greatly accelerate AI and automation among companies that survive, while increasing the frustration and anger of those left out in the cold.

The report’s authors offer suggestions to prepare for the job market of 2030, to cope with any of the above scenarios:

• Start small and learn from experience. Run small, controlled experiments to “learn from failure at low cost” and understand different technology use cases.”
• Encourage human-AI collaboration, not human replacement. This is critical for increasing trust, productivity, adoption and resilience, the researchers urge.” Skill development needs to be part of this equation, as the transition to an AI-based economy will be painful for many. Here, AI can be employed to predict talent requirements to work with AI. Invest in training, both onsite and with education providers.
• Invest in data governance and infrastructure. Without the right data, there is no AI, period. “AI models are only as good as the data they are trained on. Reliable data will be a critical source of corporate value, reputation and trust.”
• Open up your organizational culture. As management guru Peter Drucker once said, “culture eats strategy for breakfast.” That includes AI strategies by the way. Without an adaptive, forward-looking culture, you’re just dropping expensive technology into the void. “Curiosity, agility, and experimentation are the most important parts of a successful AI strategy for the future.
• Remember, no two AI stories are the same. “The pace and scale of impact from AI advancement will vary widely across occupations, tasks, geographies and sectors,” the researchers point out. “Although many routine, administrative and basic analytical tasks may face the highest early-stage displacement, others may face rising exposure with the acceleration of AI capabilities. Convergence of AI and robotics is a critical uncertainty that may affect both blue- and white-collar workers.”