Artificial intelligence has not demonstrably improved productivity metrics, according to J. P. Gownder, a vice president and principal analyst at Forrester, who stated that current AI applications are not showing up in productivity statistics.
Gownder told The Register that US Bureau of Labor Statistics data indicate past technology introductions, such as personal computers, did not immediately correlate with productivity gains. Annual productivity growth registered 2.7% from 1947 to 1973, then declined to 2.1% between 1990 and 2001, and further to 1.5% from 2007 to 2019. He noted that information technology’s impact on productivity has not always been linear. Economist Robert Solow’s 1987 observation, known as the Solow Paradox, stated that the effects of the PC revolution were visible everywhere except in productivity statistics; Gownder said this holds true for AI today.
Forrester’s recent research on AI job replacement projects that AI could displace 6% of jobs by 2030, totaling approximately 10.4 million positions. This impact stems from robotic process automation, business process automation, physical robotics, and generative AI. Gownder indicated these job losses would be structural and permanent, unlike typical post-recession job recoveries.
To assess job susceptibility, Gownder and his team analyzed about 800 job types and 34 skills defined by the US Bureau of Labor Statistics, consulting with 200 companies. Their methodology resembled the one used by University of Oxford scholars Carl Benedikt Frey and Michael Osborne in their 2013 study on job computerization susceptibility. This allowed Forrester to calculate the “automation potential” across various jobs by cross-referencing AI capabilities with identified tasks and job categories.
Gownder also discussed the effectiveness of AI implementations within large organizations, noting that “a lot of generative AI stuff isn’t really working.” He cited an MIT study indicating 95% of generative AI projects have not yielded tangible profit and loss benefits, translating to no actual return on investment. McKinsey data showed similar findings with approximately 80% of projects failing to deliver value. These outcomes suggest AI is not yet causing widespread job displacement. He clarified that recent large-scale job cuts were primarily financial decisions, not AI-driven, though some companies are delaying hiring for open roles to evaluate AI’s potential to take over those tasks.
Gownder further suggested that, historically, job losses in sectors like US manufacturing were often driven by globalization, not solely robotics. He sees a parallel with AI, where outsourcing, due to cheaper labor, can sometimes be misattributed as an AI-driven job loss.
