AI is going to eliminate all remote jobs, work from home will no longer exist: Google DeepMind co-founder
Google DeepMind co-founder Shane Legg has warned that as AI approaches human-level intelligence, remote and work-from-home roles based solely on cognitive tasks could be among the first to face disruption.
The rapid rise of artificial intelligence is no longer just a conversation about smarter tools or better productivity software. According to Shane Legg, chief scientist for AGI and co-founder of Google DeepMind, the changes coming over the next decade could be so profound as to redefine the way people work, earn, and even justify their place in the economy. In a wide-ranging interview with Professor Hannah Fry, Legg explains why AI is moving beyond assistance and into an area that could quietly wipe out large portions of remote, cognitive work.
Legg believes that human intelligence is far from the upper limit of what machines can achieve. When asked whether human intelligence sets any limits, he said, “I think not at all.” With data centers capable of consuming megawatts of power, operating at almost light speed, and processing information on a scale impossible for the human brain, he argued that machines are structurally positioned to go far beyond us. He warned that this change is no longer theoretical.
According to Legg, AI systems are already outperforming humans in many areas such as language use and general knowledge. Over the next few years, their weaknesses, including reasoning, visual understanding, and continuous learning, are likely to diminish. “My hope is that all of these things will be addressed over several years,” he said. He said AI will continue to advance toward professional-level capabilities and in areas such as coding, mathematics, and complex knowledge tasks.
This is where the future of remote jobs begins to look tenuous. Legg points out that work done entirely online, using cognitive skills rather than physical presence, is the most exposed. He offered a clear rule: “If you can work remotely over the Internet using only a computer, that job is potentially at risk.” As AI becomes more reliable and capable, companies will no longer need large distributed teams performing the same digital tasks from home.
In software engineering, he expects the change to be particularly visible. Legge suggested that teams of 100 engineers could shrink dramatically as AI takes on more of the workload. “In a few years, where before you needed 100 software engineers, you’ll probably need 20, and those 20 will be using advanced AI tools,” he said. While this increases productivity, it also means fewer roles, especially entry-level and remote roles.
However, the broader concern goes beyond specific businesses. Legge returned repeatedly to the idea that AI would “structurally transform the economy and society.” The traditional system, where people exchange mental or physical labor for income, can no longer work the way it did when machines can perform a significant part of the cognitive work better, faster and cheaper. “The current system where people contribute their mental and physical labor in exchange for access to resources can no longer work as it used to,” he said.
He acknowledged that this change would not happen overnight. Over the next few years, AI will transform from being a helpful tool to performing economically valuable tasks independently. The disruption will be uneven, with some areas hit harder than others. Remote roles that rely heavily on digital output are likely to feel the pressure already, while jobs requiring physical presence or manual skills, such as plumbing, may remain relatively safe for longer.
Legg also warned that ignoring these signs would be a mistake. He compared the current moment to the beginning of 2020, when experts warned about an impending pandemic, but many people struggled to take it seriously. “It’s very difficult for people to believe that a major change is actually coming,” he said. He said that when fundamental forces come into play, eventually major changes occur.
Despite the harsh outlook for many jobs, Legg does not see the future as entirely negative. He believes AI could open up a “real golden age” by massively increasing productivity, advancing science, and removing people from work that machines can do better. He argued that the real challenge lies in distribution, how societies share the wealth created by intelligent machines and ensure that people are not left without purpose or support.
For now, his message is clear. As AI moves closer to general intelligence, remote and work-from-home roles based solely on cognitive labor may not survive in their current form. The big question is not whether change is coming, but whether governments, institutions and industries are prepared to rethink how they work before imposing change on them.
