Elon’s & Kurzweil’s AI Predictions

What happens when AI surpasses human-level intelligence? And WHEN exactly is this likely to happen?

That’s the focus of the next Metatrend in this Age of Abundance series. 

Human-level AI, often referred to as AGI (artificial general intelligence) or ASI (artificial super intelligence), has historically been defined as the ability of a machine program to pass the “Turing Test,” defined as the ability of an AI to perform human-level tasks in a fashion indistinguishable from us humans. This definition is no longer useful.

For all intents and purposes, the Turing Test was passed, unceremoniously, sometime in the past 12 months. But no one is claiming we’ve achieved AGI or ASI yet.

Some, including futurist Ray Kurzweil and technologist Elon Musk, have argued that we may witness the emergence of true human-level AI before the end of this decade. Such powerful technology will allow any individual with an internet connection the ability to amplify their creativity, improve their problem-solving skills, and increase their earning capacity. 

Examples include OpenAI’s GPT-4 (and soon GPT-5), Google DeepMind’s recently released Gemini model, and the thousands of other large language models (LLMs) being developed around the world. 

And we’re just getting started.  

The arrival of AGI/ASI will impact every industry: from healthcare and education to advertising and finance. It will alter how we raise our kids and lead our nations.

In today’s blog, I’ll summarize Ray’s and Elon’s predictions for the development of AGI/ASI. We’ll begin with a background story on Ray, who’s a dear friend, my Co-Founder of Singularity University, and one of our many AI experts who will be at the Abundance Summit this coming March 2024.

Let’s dive in… 

A Futurist Named Ray Kurzweil

In the annals of innovation, few names shine as brightly as Ray Kurzweil. An inventor and futurist, Ray's life has been a relentless pursuit of the future, marked by pioneering inventions and predictions that have captured the world's imagination.

Born in Queens, New York, on February 12, 1948, Ray's passion for technology was ignited early. By the age of 15, he had already designed a computer program capable of composing original music, earning him an appearance on the television show I've Got a Secret. 

This remarkable achievement was a harbinger of what was to come.

Called "the restless genius" by The Wall Street Journal and "the ultimate thinking machine" by Forbes, Ray's early years were filled with invention and discovery. He was the principal inventor of the first CCD flat-bed scanner, the first omni-font optical character recognition, the first print-to-speech reading machine for the blind, and the first music synthesizer capable of recreating grand orchestral instruments.

His creations didn't just push boundaries—they shattered them. 

Ray's work was revolutionary, earning him recognition as one of the top entrepreneurs by Inc. Magazine and as the "rightful heir to Thomas Edison." His name was etched alongside the "sixteen revolutionaries who made America" by PBS.

A recipient of the National Medal of Technology and a Grammy Award for outstanding achievements in music technology, Ray was inducted into the National Inventors Hall of Fame and received honors from three U.S. presidents. His twenty-one honorary Doctorates bear testament to a mind that never ceased to innovate and inspire.

In 2012, Ray joined Google as Director of Engineering, continuing his work on machine learning and natural language processing. Today, he serves as principal researcher and “AI visionary at Google.”

Ray's written works became intellectual landmarks. His New York Times bestsellers The Singularity Is Near (2005) and How To Create A Mind (2012) became definitive texts in the realm of artificial intelligence and exponential technologies. His next book, The Singularity is Nearer, is expected in mid-2024.

To date, his written predictions, 147 in total, have demonstrated an astounding 86% accuracy (see his Wikipedia entry for more details on his predictions).

Which brings us back to his famous prediction about when exactly AI might reach and exceed human-level intelligence…

Ray’s Prediction on Human-Level AI

At a recent Abundance Summit AMA, Ray reiterated his now famous prediction: “Artificial intelligence will achieve human-level intelligence by 2029.”

“I first made my prediction on AI achieving human-level intelligence back in 1999, some twenty-four years ago. It was such a bold and alarming prediction that Stanford organized an international conference to discuss it the following year,” Ray pointed out.

During the conference, a poll was taken of global AI experts asking them if and when the attendees thought AI would achieve human-level intelligence.

“The results were incredibly conservative,” recounted Ray. “Of the roughly 150 people at the conference, 20% felt it would never happen, and 80% felt that it would happen, but not by 2029, with the majority of them predicting it would take at least 100 years.”

Over the years that followed, as AI has made continuous and accelerating progress, experts’ opinions began shifting.

“In 2021,” Ray continued, “the online forecasting platform Metaculus conducted another poll of AI experts, and the consensus was that AI would achieve human-level intelligence by 2042.”

When asked whether Ray would be changing his prediction as we have gotten closer to the 2029 date, his answer was: “No, I’m sticking with my prediction of 2029.”

But in 2022, another poll was taken among AI experts asking the same question. This time the group reached consensus around a target date of 2030—essentially in line with Ray’s original prediction that he’s stood by since 1999.

Elon’s Prediction

Elon Musk is another tech entrepreneur who makes a lot of predictions on topics like self-driving cars and landing on Mars. He’s also made predictions on when AI might achieve human-level intelligence.

In July of 2020, in an interview with the New York Times, Elon said the following on the topic: “AI will become vastly smarter than any human and would overtake us by 2025. But that doesn’t mean that everything goes to hell in five years. It just means that things get unstable or weird.”

Elon’s prediction was updated in July 2023, during a Twitter/X Spaces discussion about the future of AI, “I think it's 5 or 6 years away … and I would say the definition of digital super intelligence is that it's smarter than any human at anything.”

That would put Elon’s prediction for AGI (artificial general intelligence) directly in line with Ray’s prediction of 2029. 

But what about digital super intelligence?

As Elon has commented, “ChatGPT is well past the Turing Test …. And we are well on our way to digital super intelligence.” But he makes an important caveat about what exactly “super intelligence” might mean and what we’re up against. 

A super-intelligent AI might be smarter than a given individual human, but that doesn’t necessarily mean it will be “smarter than the sum of all humans.” That’s a higher bar. And that bar rises further once we consider that a super-intelligent AI would be competing against not the sum of all humans as they exist today, but the sum of all machine-augmented humans.

And although Elon has previously voiced his concerns about the potential threats we face with super-intelligent AI, he is doing his part to point us to a more hopeful and positive future by launching x.AI, an AI startup that will be “pro-humanity.” 

As Elon described the company’s approach, “From an AI safety standpoint … a maximally curious AI, one that is trying to understand the universe, is I think going to be pro-humanity,” rather than having an AI with moral guidelines programmed into it. 

Why This Matters

Instead of simply fearing the continued progress of AI, we should see the technology as a collaborator.

Humans have long demonstrated a remarkable ability to incorporate technological aids into new opportunities for growth. History paints a clear picture of this.

AI is no different.

In our next blog, we’ll continue the Metatrend that AI Will Achieve Human-Level Intelligence and look at several examples of what I call “Proof of AI Domination.”