"Work May Become Optional" vs "Over One Million Could Die": Experts Predict Two Possible Futures for AI
AI May Enable "Universal High Income" by Replacing Labor
AI Projected to Surpass Total Human Intelligence Within 5 Years, Reshaping Work
Experts Estimate Catastrophic AI Risk as High as 21.5%
There is a forecast that artificial intelligence (AI) could surpass the combined intelligence of all humanity within the next five years and even change the very meaning of labor and money. At the same time, experts have issued warnings that there is more than a 20% chance that AI could cause large-scale human casualties and economic losses during the same period. Expectations that AI will usher in an era of abundance for humanity are growing alongside concerns that it could also bring about an uncontrollable catastrophe.
Artificial intelligence (AI) is forecasted to surpass the combined intelligence of all humanity within the next five years, potentially transforming the very meanings of labor and money. However, experts have also warned that during the same period, AI poses more than a 20% chance of causing large-scale casualties and economic losses. Photo is unrelated to the specific content of the article. Pixabay
View original imageOn July 29, Yonhap News reported—citing the U.S. online media outlet Axios—that the MIT FutureTech interdisciplinary research group and a team at the University of Queensland in Australia announced the results of a study evaluating the risks of AI. The study surveyed 272 AI experts from 37 countries worldwide. The researchers conducted three rounds of Delphi surveys in late 2025, analyzing the likelihood and severity of 24 different types of AI-related risks.
Catastrophic AI Risk Estimated at 21.5%
First, the researchers defined a "catastrophic outcome" as a scenario involving more than 1 million deaths, over 100 billion dollars in economic losses, or intangible damage on the scale of civilization, such as the collapse of democracy and civil rights. If the current pace of AI development and tech company deployment continues, the probability of a catastrophic outcome exceeding 10% between 2025 and 2030 was estimated at 18 out of 24 risk categories. However, the researchers explained that since each risk could overlap or influence others, these probabilities should not simply be added together.
The appearance of a humanoid robot from Openmind Robotics. Photo by AFP Yonhap News
View original imageThe item with the highest estimated probability for catastrophe was "AI possessing dangerous capabilities that could escape human control or be exploited for large-scale harm," at 21.5%. This was followed by the development and use of cyberattacks and chemical or biological weapons at 21.0%. Next came the concentration of power and unfair distribution of AI benefits to a handful of companies or countries at 18.0%, competitive dynamics prioritizing performance and speed over safety at 16.6%, and the spread of misinformation and disinformation at 12.8%. Even when accounting for practical and cost-effective safety measures, five categories—including dangerous AI capabilities, weapons and cyberattacks, environmental damage, inequality and unemployment, and concentration of power—all still exceeded a 10% probability of catastrophe.
Experts assessed that AI users and the general public are the most vulnerable to these damages, but that the responsibility for risk reduction lies primarily with general-purpose AI developers, governments, and regulatory agencies. They also identified information and communications, finance, and national security as the sectors most exposed to AI-related risks. Amid these findings, Elon Musk, CEO of Tesla, projected that AI will surpass human capabilities at a much faster rate than previously expected.
Musk: "AI Will Surpass Total Human Intelligence Within 5 Years"
Musk also recently predicted in an interview with the British news magazine The Economist that within about five years, AI’s intelligence could surpass the combined intelligence of all humanity. He argued that, aside from the very existence of humans themselves, there will virtually be no tasks left that AI cannot perform better than people. If AI systems come to handle digital work such as software development and humanoid robots begin to undertake manufacturing and logistics, he said, human work could shift from an obligation for survival to a matter of personal choice.
Musk also recently predicted in an interview with the British newsweekly The Economist that AI intelligence could surpass the combined intelligence of all humanity within about five years. He claimed that, except for the very existence of humans, there would virtually be no tasks that AI could not perform better than humans. Reuters Yonhap News
View original imageMusk described this as an "era of astonishing abundance." He explained that if AI and robots produce goods and services in vast quantities and supply shortages disappear, the meaning of money could be greatly diminished within 10 years. In this process, he foresaw that governments might need to introduce a system of "universal high income" to distribute the wealth generated by automation to the public.
However, Musk also maintained his previous position, estimating the possibility of AI leading to human catastrophe at about 10 to 20%. He likened the intelligence gap between AI and humans to that between humans and chimpanzees, acknowledging that human control over advanced AI could weaken.
Hot Picks Today
"Two Spend 500,000 Won in a Flash, Yet Wait Two Hours?" ... 'Dreamland' Shunned by Kids and Foreign Tourists
- The Era of Getting 200 Million Won After 30 Years Is Over... A Japanese Bank's "Perpetual Department Head" Who Just Clocked In Cries Out
- Spain Hit by Its Worst Tragedy... Lee Kangin, Moreno and Real Madrid Join Soccer World in Mourning
- They Said, “Make It Your Home for Life” - Then Suddenly Kicked 200 Passengers Off... What Happened to the Around-the-World Cruise?
- "130 Rice Bowls Sold Out in a Day"... 'Unfiltered Face' Adds 8 Million Won to Night Market Sales: China's Bungeunwoo
Nevertheless, he argued that AI development is an irreversible trend and that it is more important to guide technological advancement in a safer direction than to attempt halting it. Musk believes the potential for AI to provide abundance for all is greater than the risk of catastrophe. He proposed that major AI companies should meet every few weeks to discuss safety and security issues and that companies should mutually verify the risks of cutting-edge models before launch. If a serious problem is discovered and the company concerned fails to address it, he suggested that government intervention would be necessary.
© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.