The rapid rise of artificial intelligence (AI) has sparked both excitement and apprehension about its impact on the workforce. In a recent interview shared on X by @slow_developer, Perplexity AI CEO Aravind Srinivas warned of a “dystopian” future where AI, while creating new value and streamlining software development, will inevitably displace jobs by making work faster and requiring fewer people. “Those who use AI well will win, but big changes in jobs and companies are coming, and no one knows how it’ll end,” Srinivas stated.
However, a 2024 study from MIT’s FutureTech offers a more measured perspective. Titled “Which Tasks are Cost-Effective to Automate With Computer Vision?”, the study conducted by researchers from MIT, The Productivity Institute, and IBM examined the economic feasibility of automating vision-related tasks, such as quality control in manufacturing or object recognition in autonomous systems. The findings challenge the narrative of rapid, widespread job displacement: only 23% of wages paid for vision-related tasks are currently cost-effective to automate. This suggests that economic barriers will slow AI adoption, leading to a “substantial but gradual” impact on the workforce rather than the mass displacement Srinivas fears.
The MIT study’s implications resonate with some reactions to Srinivas’ comments on X. User @d_symbox noted that while AI may cut jobs, it also creates new value, and those who integrate it wisely will “redefine success.”
Similarly, @ionet suggested that “small, nimble, creative teams who can leverage AI will be the future of software,” highlighting the potential for AI to augment rather than replace human work. However, concerns about the timeline persist, @ljorjoliani0 asked whether these changes might unfold in 2, 5, or 10 years, a question the MIT study indirectly addresses by emphasizing a slower adoption curve.
Broader data provides further context. According to seo.ai, 14% of workers have already experienced job displacement due to AI as of early 2025, with 35% of businesses integrating AI and 42% exploring its adoption. Yet, the MIT study suggests there’s time to adapt. Its focus on economic feasibility underscores that automation isn’t just a technological question but a financial one, offering a window for reskilling and workforce planning.
Srinivas’ warning and the MIT study together paint a complex picture: AI’s potential to disrupt is real, but its pace may be slower than feared. As @rethynkai predicted on X, large-scale displacement might hit fields like art and commerce first, while new opportunities emerge in AI and robotics. For now, the future of work seems less like a dystopian cliff and more like a gradual slope one that demands adaptation but offers time to prepare.