Feature not a bug: Anthropic CEO Dario Amodei says AI could cause significant enduring job loss

Anthropic chief Dario Amodei has renewed his warning that AI could cause lasting job losses and urged governments to act. He says the technology's gains could be vast, but institutions must prepare faster to protect workers and spread the benefits.

In Short

• Dario Amodei maintains his warnings about AI job losses
• Anthropic CEO proposes measures to help affected workers
• Says governments must prepare for rapid AI disruption

OpenAI CEO Sam Altman has recently appeared to soften his stance on AI-driven job losses, increasingly suggesting that some of his earlier predictions may have been overstated. But his biggest rival, Anthropic CEO Dario Amodei, is not having a similar change of heart. Instead, Amodei is doubling down on his warnings about the impact of artificial intelligence on jobs. More importantly, he is increasingly focusing on what governments can do to reduce the damage and ensure that the benefits of AI are shared more broadly.

The Anthropic CEO recently published an essay titled "Policy on the AI Exponential." In the essay, Amodei argues that there is a "decent possibility" that, despite efforts to soften the blow, AI could cause "significant enduring job loss" and that this "may be an intrinsic property of the technology and the way it broadly replicates human cognition."

In other words, Amodei argues that widespread job displacement may not simply be an unintended side effect of AI. According to him, it could stem from the very nature of the technology, which is designed to replicate human cognition and perform tasks that would otherwise require human thinking.

Amodei has sounded alarms on this before. Earlier this year, he warned that AI could eliminate half of entry-level white-collar jobs within five years and push unemployment to between 10 per cent and 20 per cent. At the time, he urged companies and policymakers to stop "sugarcoating" the risks.

His latest essay expands on those concerns. Amodei argues that AI is advancing far faster than governments and institutions can respond. He says society is approaching an era of "Powerful AI", systems capable of performing many cognitive tasks at or above the level of human experts, and that policymakers need to act now rather than later.

Amodei warns that it may replace human cognitive work on a much broader scale than previous technological revolutions, potentially leading to long-lasting job displacement rather than temporary disruption.

How Amodei wants governments to respond

Rather than focusing only on the risks, Amodei outlines several measures that governments could take to prepare for AI-driven disruption. He called for better measurement and tracking of AI's impact on the labour market, including expanded government statistics. He also endorsed what he described as "pro-employment incentives," including wage insurance for workers who are forced to take lower-paying jobs because of AI, retention tax incentives, workforce training grants, and improved job-matching infrastructure.

The goal, according to Amodei, is to help workers adapt to technological change while ensuring that the gains from AI are distributed more widely.

Not predicting doom, but urging preparation

Despite his warnings, Amodei is not calling for AI development to be slowed down, nor is he arguing that catastrophe is inevitable. Instead, he says AI's benefits could be enormous, but governments, businesses, and societies need to adapt much faster than they are today.

Speaking recently to Bloomberg, Amodei responded to critics who suggested that his warnings about AI risks were simply a marketing strategy. He rejected that idea and said his message is not that "doom is coming," but rather that society should recognise the potential risks early and prepare for them before they become harder to manage.

- Ends