Artificial intelligence may have crossed a technological threshold popularly known as “Singularity,” according to one expert who says we have surpassed the point where machines merely assist us, with intelligent machines now exceeding human capabilities.
The claims were made by OpenAI CEO Sam Altman, who in a recent blog post wrote that AI is already demonstrating intelligence that goes beyond what humans can do naturally, but that for now, things aren’t looking as strange as some might have expected once we reached this point.
“We are past the event horizon; the takeoff has started,” Altman wrote. “Humanity is close to building digital superintelligence, and at least so far it’s much less weird than it seems like it should be.”
The Singularity is Here
For decades, futurists have predicted the arrival of a point where technological advancements would reach a point of no return, potentially instigating a transformation of human civilization. Popularly known as the “Singularity,” definitions may vary, although most visions of this technological event horizon have one thing in common: the rise of artificial intelligence.
Only a few years ago, the idea that AI could mimic the most demanding human tasks would have seemed like a remote, futuristic possibility. However, that has all changed with the rise of chatbots like ChatGPT and others, which are increasingly pushing the boundaries of machine intelligence into unforeseen territory.
Recent advancements in AI have already shown that machine intelligence is evolving in ways that make its performance more like human thought. This allows it to be trained to solve problems and make decisions more like humans while limiting the potential for “hallucinations.”
Going beyond simply mimicking the most useful functions of human thought, it was reported in April that an AI developed by the Artificial Scientist Lab at the Max Planck Institute for the Science of Light had successfully developed new concepts for gravitational wave detectors, some of which scientists still don’t fully understand.
The Era of Superintelligence
For Altman and OpenAI, the age of digital superintelligence is no longer speculative—it has already arrived, although it doesn’t look quite as leading futurists might have expected.
“Robots are not yet walking the streets, nor are most of us talking to AI all day,” Altman said. “People still die of disease, we still can’t easily go to space, and there is a lot about the universe we don’t understand.”
“And yet, we have recently built systems that are smarter than people in many ways,” Altman argues, which he says are already greatly increasing the output of those who use these innovative AI models. To Altman, the most challenging elements of the work required to achieve true superintelligence have already been surpassed. Now we are looking ahead at what creations like his company’s famous ChatGPT may bring us in the years ahead.
“In some big sense, ChatGPT is already more powerful than any human who has ever lived,” Altman wrote, emphasizing the large numbers of individuals who rely on it daily, and that this year has also marked the rise of AI “agents” which he says “can do real cognitive work,” fundamentally changing the field of computer programming and coding technologies.
The resulting feedback loops extend beyond software and coding, however. Economic gains are fueling large-scale infrastructure that includes data centers, chip factories, and robotics. To Altman, it is only a matter of time before robotics, fueled by AI, will be capable of building new robots, as new AI-driven data systems will expand the growth of future data centers.
Into the 2030s, and Beyond
Looking ahead to the 2030s, Altman predicts that advancements we are currently seeing will lead to even more fundamental shifts in society in the coming decade. Eventually, as automation increases, Altman argues that the cost of intelligence could conceivably even converge with the cost of electricity.
“Intelligence too cheap to meter is well within grasp,” Altman wrote.
Of course, concepts like self-replicating robots and data centers, solving high-energy physics problems in ways that could significantly accelerate the path toward distant space travel, major materials science breakthroughs that make emerging technologies like brain-computer interfaces common aspects of our everyday lives do probably sound ambitious, at very least—perhaps even unattainable for some.
Altman, however, remains optimistic, especially in light of how quickly AI like ChatGPT and its numerous competitors have emerged within a short period of time, quickly becoming assimilated into our lives at multiple different levels.
“This may sound crazy to say, but if we told you back in 2020 we were going to be where we are today, it probably sounded more crazy than our current predictions about 2030,” Altman says.
Micah Hanks is the Editor-in-Chief and Co-Founder of The Debrief. He can be reached by email at micah@thedebrief.org. Follow his work at micahhanks.com and on X: @MicahHanks.