There is a 50/50 chance AI will take your job

Imagine logging into your work computer to find that half of your daily tasks have been seamlessly automated overnight. Your email responses are drafted, your data is analyzed, and your reports are summarized — all by artificial intelligence. Science fiction? 

Far from it. A study published in Science in June, titled “GPTs are GPTs: Labor market impact potential of LLMs,” suggests this scenario is not just possible — it is on the horizon for nearly half of us.

Researchers from OpenAI, the Centre for the Governance of AI, and the Wharton School examined our AI-driven near future, and the view is startling. Their findings? About 1.8 percent of jobs could see over half their tasks affected by AI using current large language models. These include OpenAI, ChatGPT, Anthropic’s Claude and Google’s Gemini, using simple chatbot interfaces. 

But here is the kicker: When accounting for likely near-term software developments that complement LLM capabilities, this jumps to a staggering 46 percent. These developments include specialized software that can interface between LLMs and domain-specific tools, AI-powered coding assistants to generate and debug complex software, and advanced data analysis tools that interpret and visualize large datasets based on natural language prompts.

This is not about some far-off future or hypothetical breakthroughs. We are discussing integrating existing AI technologies with software developments already in progress or on the immediate horizon. The scale of potential disruption is not just possible — it is probable.

Surprisingly, it is not always the jobs you might expect that are most exposed. The study found that occupations requiring extensive preparation — think lawyers, pharmacists and database administrators — are more exposed to AI than those with lower entry barriers. This challenges the common perception that AI primarily threatens people in low-skilled jobs.

However, “exposure” does not necessarily mean job losses. The study emphasizes that AI’s impact is more about transformation than replacement. When we say a job is “exposed” to AI, we mean that many tasks within that job may be augmented or automated. This often leads to shifts in job roles rather than outright elimination. The critical question is how quickly these changes will occur and how prepared we are to adapt.

For workers, the imperative is clear: Adaptability is crucial. Regardless of your field, developing skills that complement AI rather than compete with it will be essential. This might involve focusing on uniquely human capabilities like creativity, emotional intelligence, and complex problem-solving.

How we develop, deploy, and regulate AI technologies in the immediate future will shape their impact on the workforce.

Mohammed A. Alqarni

Businesses need to start planning for an AI-augmented workforce now, not in some distant future. This does not just mean adopting new technologies but rethinking entire business processes and job roles to leverage the strengths of humans and AI.

Policymakers face an urgent challenge. The study calls for immediate public investment in tracking AI adoption and its labor market impacts. We need robust policies to ensure the benefits of AI are broadly distributed and to support workers through what could be rapid transitions. This might include reforms in education, upskilling and reskilling programs, and social safety nets.

The “50/50” nature of AI’s potential impact on jobs underscores a crucial point: While significant disruption appears likely, the specific outcomes are not predetermined. How we develop, deploy, and regulate AI technologies in the immediate future will shape their impact on the workforce. We have the power — and the responsibility — to influence this future, starting right now.

As millions of us face this looming transformation in our working lives, it is clear that we need to rethink the future of work fundamentally. This is not just about adapting to AI — it is about reimagining what work means in an AI-augmented world. We must ask ourselves: How can we redefine productivity and value creation? What new forms of work might emerge? How can we ensure that the benefits of AI are distributed equitably across society?

This rethinking needs to happen at all levels. Individuals should consider how their roles might evolve and what skills they will need in an AI-rich environment. Companies must reevaluate their organizational structures and how they can best leverage human-AI collaboration. Policymakers and educators must overhaul our education and training systems to prepare people for this new world of work.

The question is not whether AI will change our jobs but how quickly and dramatically. Our challenge — and our opportunity — is to harness its imminent potential to create a more productive, equitable, and fulfilling world of work for all. This requires not just adaptation but a bold reimagining of “work” itself. The time for this rethinking is now before the change overtakes us. 

Let us seize this moment to shape a future of work that benefits everyone, putting human needs and aspirations at the center of our AI-augmented future.

• Mohammed A. Alqarni is an academic and AI business consultant