While debate over the timeline – or even the potential – for artificial general intelligence (AGI) rages on in 2026, one futurist may have predicted the breakthrough more than 60 years ago.
Noted British science fiction writer and futurist Arthur C. Clarke touted the arrival of AGI during an interview at the 1964 World’s Fair in New York City.
Article continues below
“Present-day electronic brains are complete morons, but this will not be true in another generation. They will start to think, and eventually, they will completely out-think their makers.”
Clarke pondered whether this prospect was “depressing”, but noted that advances in technology on this front represent the next evolutionary step in humanity’s journey.
“We superseded the Cro-Magnon and Neanderthal men, and we presume we’re an improvement,” he added.
“We should regard it as a privilege to be stepping stones to higher things. I suspect that organic evolution has about come to its end, and we are now at the beginning of inorganic or mechanical evolution, which will be thousands of times swifter.”
The AGI conundrum
The debate over whether AGI is even attainable has raged for some. While typically confined to the realms of science fiction, the advent of generative AI in late 2022 once again brought the topic back to the fore.
Notably, debate over the actual definition of AGI is a key sticking point for many in the industry, and society more broadly. By Google’s definition, for example, AGI refers to:
“The hypothetical intelligence of a machine that possesses the ability to understand or learn any intellectual task that a human being can. It is a type of artificial intelligence (AI) that aims to mimic the cognitive abilities of the human brain.”
Taking this into account, it’s safe to say that humanity hasn’t reached AGI quite yet, or is anywhere close to reaching that goal. But major industry players such as OpenAI insist that reaching AGI is their ultimate end goal.
In a 2025 blog post last year, OpenAI CEO Sam Altman reflected on the company’s pursuit of this so far elusive moment, noting that progress is being made at a nominal pace.
“We are now confident we know how to build AGI as we have traditionally understood it,” Altman wrote.
During a September 2025 interview at the WELT AI Summit, Altman once again banged the drum for an imminent AGI breakthrough, claiming that AI will surpass human intelligence by 2030.
It’s worth noting that OpenAI’s own definition of AGI differs from that of the aforementioned Google – a fact that underscores the conflicting outlook on this subject.
OpenAI defines AGI as a “highly autonomous system that outperforms humans at most economically valuable work”. That definition might, at least in some circles, be a rather low bar to set – especially given advances in AI over the last 18 months.
The path to AGI is becoming clearer
The advent of agentic AI suggests that progress by OpenAI’s definition is being made to some extent. Rather than traditional AI “assistants” rolled out by big tech providers during the early days of the generative AI boom, agents are capable of autonomously conducting tasks on behalf of human workers.
That marked a step change in how enterprises and consumers alike engage with the technology, and it has wide-reaching implications for the future of work. Areas such as customer service, for example, have been firmly in the crosshairs of agentic AI providers, with these roles identified as prime candidates for automation.
In other professions, such as software development, AI is already outperforming human workers in areas such as coding.
To some, these advances might point toward humanity reaching the tipping point on AGI, but a key factor in whether or not AGI can be acknowledged lies in generality.
Specialist AI tools or agents aimed specifically at conducting one particular task isn’t a marker of AGI, more that these tools and bots have been trained with these tasks in mind.
However, being able to switch between tasks and carry them out at the same level of efficiency is, according to Google. Core characteristics of AGI by the tech giant’s definition include “generalization ability”.
“AGI can transfer knowledge and skills learned in one domain to another, enabling it to adapt to new and unseen situations effectively,” the company notes.
Common cause in big tech
Altman isn’t the only leading industry figure convinced that AGI is achievable and looming around the corner. A host of industry leaders such as Dario Amodei and Elon Musk have also touted the potential in the near future.
What these figures all have in common, however, is that their long-term roadmap is based on achieving this goal, and it’s becoming increasingly important in deals between industry players.
OpenAI’s revised partnership agreement with Microsoft, for example, contains clauses relating to AGI. A similar partnership with Amazon Web Services (AWS) also requires the company to reach AGI to secure future investment boosts.
Clarke may have loosely predicted a future of intelligent machines capable of human-level thinking, but what he likely couldn’t have predicted is exactly how much was at stake from a financial perspective.
https://cdn.mos.cms.futurecdn.net/rNZmVCrdHzszaDCyTrBWmj-2560-80.jpg
Source link
