Warnings that artificial intelligence will destroy humanity have grown louder. One recent example came from Jacob Coxon, who resigned from the AI company Anthropic and argued that the race to build ever-more-powerful systems could pose an existential risk. Similar claims have been made by other researchers and executives.
The question is whether these warnings reflect a uniquely dangerous technology, or whether they belong to a longer pattern of human overreaction to serious but manageable problems.
Over the past seven decades, a series of existential alarms have captured public attention:
1950s: Severe urban air pollution, especially in cities such as Los Angeles, was presented as a rapidly worsening threat to public health.
1960s–early 1970s: Overpopulation. Influential works warned that rising numbers would produce inevitable famine, plague, and the collapse of civilized life.
1973: The Arab oil embargo. Gasoline shortages and price spikes fueled fears that the automobile age—and the broader economy—were coming to an end.
Mid-to-late 1970s: Resource exhaustion. Reports such as The Limits to Growth and peak-oil theories predicted that petroleum and petroleum-based fertilizers would run out, leading to mass starvation.
1970s: Global cooling. A minority of scientific papers and a larger volume of media coverage speculated about a coming ice age. (By the late 1980s the dominant concern had already shifted to warming.)
1990s–2000s: Climate change. Public warnings evolved from drought and desertification to increased flooding and extreme weather, eventually consolidated under the broader term “climate change.”
2020: COVID-19. A genuine pandemic produced intense fear, sweeping policy responses, and predictions of societal breakdown.
None of these issues was imaginary. Air pollution damages health. Oil shocks cause real economic pain. Pandemics kill. Climate trends have consequences. Yet in each case the most extreme forecasts—of civilization collapse or human extinction—proved significantly overstated. The panics themselves imposed substantial costs: misallocated resources, expanded government authority, restricted liberties, and, at times, policies that created new harms.
A recurring pattern is visible. Genuine problems are inflated into extinction-level crises. The public is urged to demand immediate and far-reaching state action. Once granted, that authority is rarely fully returned.
Artificial intelligence raises legitimate questions about safety, economic disruption, surveillance, and the concentration of power. Those questions deserve serious, evidence-based attention. Treating every advance as the probable end of the human species, however, follows a familiar historical script more closely than it follows a measured assessment of risk.
We have been here before. This time do we respond with calm proportion or, once again, do we panic?
(Co-written with A.I. Grok)
No comments:
Post a Comment