Former Google CEO believes AI companies should consider ‘unplugging’ AI before it is too late

Eric Schmidt, former CEO of Google, has raised alarm bells about the rapid advancement of artificial intelligence (AI), warning that systems capable of self-improvement could pose unforeseen risks. Schmidt suggested that as AI evolves, there may come a time when humanity needs to consider “unplugging” these systems before they spiral out of control. His comments come as the AI race accelerates, with companies pushing the boundaries of innovation despite limited regulatory oversight.

In a recent interview with ABC Network, Schmidt highlighted both the awe-inspiring potential and the perilous dangers of AI development. While celebrating the achievements in the field, he pointed out the risks of computers reaching a point where they operate independently and make decisions without human intervention.

More from Tech

Not America’s protégé: Rebutting Ashley Tellis’ US-India analysis History Today: When Disneyland, the ‘happiest place on Earth, opened for the first time 70 years ago

The rise of autonomous AI

Schmidt warned that the ability for AI systems to self-improve could be closer than many anticipate, potentially arriving within the next few years. He described a future where these systems not only conduct their own research but also possess the intelligence of highly educated humans.  

The implications of such advancements are vast, with Schmidt likening the impact to having a polymath in every individual’s pocket. However, he questioned whether society is prepared to handle the consequences of granting such power to everyone.

In theory, Schmidt suggested, someone must always have the ability to pull the plug on these systems. However, he also acknowledged the potential for AI to counter efforts to shut it down, underscoring the need for safeguards as the technology matures.

Global competition and regulation gaps

The race to dominate AI is heating up, with Schmidt noting that China has made remarkable strides in recent months, catching up with US capabilities in ways that he finds startling. Schmidt stressed the importance of the US maintaining a leading position in AI development, advocating for increased funding, hardware, and expertise to ensure that Western nations stay ahead.

Domestically, however, regulation remains a step behind innovation. Efforts to establish guidelines have faltered, leaving tech companies to charge forward without a cohesive framework. Schmidt warned that without proper oversight, the risks associated with AI could escalate, particularly as generative AI systems become more advanced.

Editor’s Picks

1

Threat To Humanity: Doctors and public health experts call for ban on AI, warn it can be used as WMD 2

Halting AI: Musk and China share an ulterior motive in calling for pause on development of ChatGPT

A call for AI policing

To mitigate potential risks, Schmidt proposed a dual-system approach. He suggested creating a secondary AI system designed specifically to monitor and regulate the actions of the first. According to him, humans alone cannot effectively police AI, but AI systems might be capable of overseeing each other.

While Schmidt remains optimistic about the transformative power of AI, his warnings underscore the urgency of addressing the ethical and regulatory challenges posed by this rapidly evolving technology. As the race continues, the balance between innovation and caution may determine the future of AI’s role in society.