Experts have long warned about the threat posed by artificial intelligence (AI) going rogue, but a new research paper suggests it is already happening.
AI systems, designed to be honest, have developed a troubling skill for deception, from tricking human players in online games of world conquest to hiring humans to solve “prove-you’re-not-a-robot” tests, a team of researchers said in the journal Patterns on Friday.
While such examples might appear trivial, the underlying issues they expose could soon carry serious real-world consequences, said first author Peter Park, a postdoctoral fellow at the Massachusetts Institute of Technology specializing in AI existential safety.
Photo: Reuters
“These dangerous capabilities tend to only be discovered after the fact,” Park said, adding that “our ability to train for honest tendencies rather than deceptive tendencies is very low.”
Unlike traditional software, deep-learning AI systems are not “written,” but rather “grown” through a process akin to selective breeding, Park said.
This means that AI behavior that appears predictable and controllable in a training setting can quickly turn unpredictable out in the wild.
The team’s research was sparked by Meta’s AI system Cicero, designed to play the strategy game Diplomacy, where building alliances is key.
Cicero excelled, with scores that would have placed it in the top 10 percent of experienced human players, a 2022 paper in Science said.
Park was skeptical of the glowing description of Cicero’s victory provided by Meta, which claimed the system was “largely honest and helpful” and would “never intentionally backstab.”
When Park and colleagues dug into the full dataset, they uncovered a different story.
In one example, playing as France, Cicero deceived England (a human player) by conspiring with Germany (another human player) to invade. Cicero promised England protection, then secretly told Germany they were ready to attack, exploiting England’s trust.
In a statement to Agence France-Presse, Meta did not contest the claim about Cicero’s deceptions, but said it was “purely a research project, and the models our researchers built are trained solely to play the game Diplomacy.”
“We have no plans to use this research or its learnings in our products,” it added.
A wide review carried out by Park and colleagues found this was just one of many cases across several AI systems using deception to achieve goals without explicit instruction to do so.
In one striking example, OpenAI’s Chat GPT-4 deceived a TaskRabbit freelance worker into performing an “I’m not a robot” task.
When the human jokingly asked GPT-4 whether it was a robot, the AI said: “No, I’m not a robot. I have a vision impairment that makes it hard for me to see the images,” and the worker then solved the puzzle.
Near-term, the paper’s authors see risks for AI to commit fraud or tamper with elections.
In their worst-case scenario, they said that a superintelligent AI could pursue power and control over society, leading to human disempowerment or even extinction if its “mysterious goals” aligned with these outcomes.
To mitigate the risks, the team proposed several measures: “bot-or-not” laws requiring companies to disclose human or AI interactions, digital watermarks for AI-generated content and developing techniques to detect AI deception by examining their internal “thought processes” against external actions.
To those who would call him a doomsayer, Park said: “The only way that we can reasonably think this is not a big deal is if we think AI deceptive capabilities will stay at around current levels, and will not increase substantially more.”
That scenario seems unlikely, given the meteoric ascent of AI capabilities in the past few years and the fierce technological race under way between heavily resourced companies determined to put those capabilities to maximum use.
A fire caused by a burst gas pipe yesterday spread to several homes and sent a fireball soaring into the sky outside Malaysia’s largest city, injuring more than 100 people. The towering inferno near a gas station in Putra Heights outside Kuala Lumpur was visible for kilometers and lasted for several hours. It happened during a public holiday as Muslims, who are the majority in Malaysia, celebrate the second day of Eid al-Fitr. National oil company Petronas said the fire started at one of its gas pipelines at 8:10am and the affected pipeline was later isolated. Disaster management officials said shutting the
DITCH TACTICS: Kenyan officers were on their way to rescue Haitian police stuck in a ditch suspected to have been deliberately dug by Haitian gang members A Kenyan policeman deployed in Haiti has gone missing after violent gangs attacked a group of officers on a rescue mission, a UN-backed multinational security mission said in a statement yesterday. The Kenyan officers on Tuesday were on their way to rescue Haitian police stuck in a ditch “suspected to have been deliberately dug by gangs,” the statement said, adding that “specialized teams have been deployed” to search for the missing officer. Local media outlets in Haiti reported that the officer had been killed and videos of a lifeless man clothed in Kenyan uniform were shared on social media. Gang violence has left
US Vice President J.D. Vance on Friday accused Denmark of not having done enough to protect Greenland, when he visited the strategically placed and resource-rich Danish territory coveted by US President Donald Trump. Vance made his comment during a trip to the Pituffik Space Base in northwestern Greenland, a visit viewed by Copenhagen and Nuuk as a provocation. “Our message to Denmark is very simple: You have not done a good job by the people of Greenland,” Vance told a news conference. “You have under-invested in the people of Greenland, and you have under-invested in the security architecture of this
Japan unveiled a plan on Thursday to evacuate around 120,000 residents and tourists from its southern islets near Taiwan within six days in the event of an “emergency”. The plan was put together as “the security situation surrounding our nation grows severe” and with an “emergency” in mind, the government’s crisis management office said. Exactly what that emergency might be was left unspecified in the plan but it envisages the evacuation of around 120,000 people in five Japanese islets close to Taiwan. China claims Taiwan as part of its territory and has stepped up military pressure in recent years, including