Bailey: AI is next GPT, growth impact uncertain
Bank of England Governor Andrew Bailey discussed the potential of Artificial Intelligence as the next General Purpose Technology. Speaking at the 389th Cutler's Feast in Sheffield, he highlighted the uncertain timeline for AI's significant contribution to productivity growth.
UK's productivity puzzle
Bailey opened by noting the significant slowdown in the UK's potential growth rate since the Financial Crisis.
From 1990-2006, average annual potential supply growth was 2.8 percent, with productivity contributing 2.4 percentage points.
Post-Crisis, this fell to 1.3 percent, with productivity's contribution dropping to 0.4 percentage points.
This decline has directly impacted living standards, with national income per head growing at 0.6 percent annually post-Crisis, compared to 2.0 percent previously.
Bailey referenced Adam Smith's division of labour and Joseph Schumpeter's creative destruction as complementary theories explaining long-run growth, emphasizing the role of institutions and trade in fostering innovation and economic development, citing the Cutler's Company's historical role.
AI: The next general purpose technology?
The Governor used Henry Bessemer's steel converter as an example of creative destruction, noting its broad impact despite initial resistance.
He distinguished General Purpose Technologies (GPTs) – such as the steam engine, electricity, and the internet – from mere products, defining GPTs as technologies with economy-wide applicability that enable complementary innovations.
Bailey argued that the recent 15-year growth slowdown could be a 'gap between cycles' of GPTs.
He identified Artificial Intelligence, particularly when combined with robotics, as the most likely next GPT.
He defined AI in terms of machines processing vast information and, crucially, learning from experience through reinforcement learning, which allows for adaptive judgement and tacit knowledge development.
Innovation's long road to productivity
Bailey's core message is that while AI can significantly improve productivity in areas like cutlery manufacturing, its broader economic impact remains an open question.
History shows a substantial time lag between the emergence of General Purpose Technologies and their measurable contribution to aggregate productivity statistics.
Policymakers must therefore manage expectations, focusing on fostering complementary innovations rather than anticipating immediate, dramatic shifts in economic growth figures.
Source: Can AI make cutlery? - speech by Andrew Bailey
IN: