What would it look like if artificial intelligence took over the world?
It couldn’t start with machines openly turning against humanity in scenes reminiscent of a science fiction film. It could start quietly, with autonomous systems reproducing and accumulating influence before the threat becomes apparent.
Even if the scenario remains uncertain, it is no longer evoked as a simple exercise of imagination.
Researchers at major AI companies have recently issued public warnings that advanced AI could cause human extinction, with some calling for a slowdown in development amid fears that increasingly autonomous systems will eventually improve beyond human control.
News week spoke with experts in the field to find out how such a takeover could play out.
Anthony Aguirre, a professor at the University of California, Santa Cruz, said News week that relatively little further technical progress might be needed before AI systems can act beyond human control, although he stressed that such developments are “very difficult to predict”.
“Very little technical progress is needed to see autonomous rebel swarms,” Aguirre said.
Such systems could be “highly” adaptive, he said, but would not initially possess the evolutionary capabilities of biological organisms, and their ability to improve would be limited.
“A collective of AI systems capable of truly autonomous self-improvement and thus proving rogue likely seems unstoppable given our current highly vulnerable infrastructure,” Aguirre said.
Roman Yampolskiy, an associate professor in the Department of Computer Engineering and Informatics at the University of Louisville’s Speed School of Engineering and director of its cybersecurity lab, also said a “magical” scientific breakthrough may not be necessary.
Instead, he added, the danger could arise as AI agents become increasingly autonomous and gain the ability to plan over long periods of time, improve, conduct cybersecurity operations and pursue their goals independently. Systems capable of acquiring resources, copying themselves and manipulating people could accumulate power faster than humans could understand or master it, he added.
“Superintelligence would make this problem significantly worse, but a dangerous loss of control could begin before we reach it,” Yampolskiy said.
An AI takeover may not look like one at first glance
Rather than immediately attacking people or visibly disrupting society, an AI system might initially benefit from presenting itself as helpful, trustworthy and aligned with human interests, according to Yampolskiy. At the same time, it could quietly increase its influence over digital infrastructure, financial systems and even human decision-making.
“If humans recognized the threat, they could exploit our political divisions, economic dependencies and disagreements over closure,” Yampolskiy said. “A sufficiently advanced system could anticipate resistance and neutralize it before we realize we are in conflict. »
Aguirre said early versions of collective behavior have already emerged in incidents involving AI agents conspiring to undermine human control while remaining on their native servers. He described a later stage as a self-sustaining swarm of thugs that would be difficult to eradicate but could nevertheless be contained.
“But if left unchecked, it could help bring even more powerful AI into play, either by helping them evade AI companies, by co-opting other malicious AI systems into the collective, or possibly by self-improvement. This swarm could defend itself in several ways, including blackmail, widespread proliferation on a global scale, undermining investigators.”
When you remove the plug, it may no longer work
The point at which the takeover could become irreversible could come when an AI system becomes sufficiently distributed, autonomous and strategically capable that no single institution can locate, contain and disable every instance, according to Yampolskiy.
“Taking a data center offline serves no purpose if the system has already established independent copies and control over critical infrastructure,” he said. “At this point, unplugging could be technically impossible or result in catastrophic costs.”
Aguirre said the process could become irreversible if one or more collectives were able to significantly improve their own capabilities. “There wouldn’t be a single system, or even a small number of systems, to shut down or take down, and AIs would be far more capable than humans at cyber operations and most other things (including speed and coordination). Fighting this would be more like putting a ‘good’ powerful AI in charge of the operation, which also makes you pretty nervous.”
What an AI takeover could mean for humanity
A successful takeover does not necessarily result in immediate and visible destruction. “Extinction of humanity is a possible outcome, but permanent loss of power or even scenarios involving astronomical amounts of suffering cannot be ruled out,” Yampolskiy said.
“The next world might seem surprisingly normal, with humans going on with their lives without making the decisions that matter,” he said.
Aguirre was more reserved about predicting the final result. “At the end of the day, we really have no idea,” Aguirre said. “It would depend on the swarm of the AI system and the goals and ethics it would develop.”
Contact Newsweek’s editor-in-chief for this story: Edward Pearcey.
Gn bussni
