MIT Study Eases Job Automation Worries: Only 23% of Tasks Cost-Effective for AI Replacement. Human Skills and Intuition, Along with AI Complexities, Resist Automation Impact Across Industries. Persistence of Human-Focused Jobs Expected Until 2042.
A recent study from the Massachusetts Institute of Technology (MIT) illuminates the practical timeline for artificial intelligence (AI) to supplant human jobs, suggesting that concerns regarding an impending robotic domination may be premature. The research, entitled “Beyond AI Exposure: Identifying Tasks Cost-Effective for Automation with Computer Vision,” carried out by five researchers from MIT, concentrated on tasks appropriate for AI, including those executed by educators, real estate appraisers, and pastry chefs.
The investigation additionally discloses that, in contrast to prevailing apprehensions, merely 23% of the earnings of employees in these occupations could be efficiently substituted by AI. Despite a 50% annual reduction in costs, the analysts forecast that it will be 2026 before half of the visual tasks attain a machine economic advantage. They expect that roles emphasizing human labor will endure until at least 2042.
Financed by the MIT-IBM Watson AI Lab, the research collected data through online surveys encompassing 1,000 “visual inspection” assignments, including tasks like examining food to determine its freshness, across 800 professions. The findings indicate that at present, merely 3% of these tasks are economically automatable. However, the researchers propose that this percentage could escalate to 40% by 2030, subject to reduced data costs and enhanced accuracy.
Why AI Will not Steal Your Job Soon
The research underscores that although AI technology excels in pattern recognition and image analysis, it entails significant installation and maintenance expenses. In numerous cases, opting for human skills and intuition proves to be a more cost-effective choice. Furthermore, AI encounters challenges related to power consumption, and organizations are contending with the intricacies of implementing AI systems.
Importantly, the research highlights that AI, propelled by statistical and symbolic reasoning, falls short of the depth inherent in human conscious and subconscious cognitive processes. Activities demanding intuition, gut feeling, and implicit knowledge—integral to human critical thinking and emotional intelligence—persistently defy automation. These qualities retain their irreplaceable value in the ever-changing job landscape.
While recognizing the influence of AI across diverse sectors such as banking, marketing, retail, legal services, transportation, and healthcare, the research asserts that most jobs face susceptibility, though not immediate vulnerability to automation. A report from Goldman Sachs in 2023 estimates that roughly 18% of global employment, equivalent to approximately 300 million jobs, might be impacted by generative AI. The MIT study acknowledges that ongoing AI advancements, including improved data efficiency and accuracy, hold the potential to substantially enhance its automation capabilities in the future.
However, at present, the narrative of “AI stealing our jobs” appears to be quite exaggerated.
