Photo by luis gomes on Pexels
As artificial intelligence increasingly shapes our world, a crucial debate persists: is AI a unique, potentially uncontrollable force, or simply another technology evolving within society? While some experts raise alarms about impending ‘superintelligence’ and demand stringent regulations, others explore AI rights and its expanding role in fields like music and therapy. This dichotomy fuels contrasting visions, ranging from a technological renaissance to widespread societal disruption.
Princeton researchers Arvind Narayanan and Sayash Kapoor offer a counterpoint, arguing that AI should be viewed as a general-purpose technology akin to electricity or the internet. In a thought-provoking essay, they emphasize the importance of understanding the often-overlooked adoption process and the time lag between technological advancement and societal integration.
Kapoor suggests that focusing on immediate, transformative consequences overlooks the nuances of how technologies are actually absorbed and utilized. Both researchers dismiss the concept of ‘superintelligence’ as speculative and ill-defined. They envision AI creating new roles centered on oversight and verification, rather than wholesale automation. Their analysis highlights AI’s potential to exacerbate existing societal challenges, such as inequality and democratic erosion, depending on how it’s implemented.
The authors note the exclusion of AI’s military applications, citing the difficulty in accessing classified data, but indicate that future research will address this area.
Treating AI as a ‘normal’ technology challenges the prevailing national security-centric approach favored by both the current and previous administrations. Narayanan critiques the US-China ‘AI arms race’ narrative as unrealistic, arguing that AI knowledge spreads quickly across borders, making secrecy unviable.
Instead of dwelling on science fiction scenarios, the researchers advocate for strengthening democratic institutions, enhancing technical expertise within government, promoting AI literacy, and encouraging AI adoption by those dedicated to its responsible and ethical use. These pragmatic proposals represent a move towards a more grounded and less sensationalized approach to AI governance.