Humanity’s millennia-long march toward the Singularity has become a sprint. — Ray Kurzweil, The Singularity is Nearer: When We Merge with AI
If Ray Kurzweil is correct, and he usually is, by 2045 human life will not exist as we know it. As he writes in his newly released book,
Eventually nanotechnology will enable [us to] expand our brains with layers of virtual neurons in the cloud. In this way we will merge with AI and augment ourselves with millions of times the computational power that our biology gave us. [Emphasis mine]
When he first made this forecast years ago many readers couldn’t believe such a fantastic claim. But in the 19 years since release of The Singularity is Near (2005), information technology is proving him right. Consider:
The Singularity Is Nea...
Best Price: $13.98
Buy New $14.00
(as of 09:52 UTC - Details)
Social media and smartphones have gone from virtually nonexistent to all-day companions that now connect a majority of the world’s population.
Algorithmic innovations and the emergence of big data have allowed AI to achieve startling breakthroughs sooner than even experts expected—from mastering games like Jeopardy! and Go to driving automobiles, writing essays, passing bar exams, and diagnosing cancer.
Now, powerful and flexible large language models like GPT-4 and Gemini can translate natural-language instructions into computer code—dramatically reducing the barrier between humans and machines.
Meanwhile, neural networks have begun simulating biology digitally, creating a path to major medical discoveries. And the cost of computing power now is roughly 11,200 times cheaper than it was in 2005.
To drive this point home, Kurzweil compares the IBM 7094 mainframe he used as an undergraduate at MIT in 1965 to Apple’s iPhone 14 Pro. In 2023 dollars the IBM mainframe cost $30 million, the iPhone $999.
At least roughly [he writes], the iPhone is 68 million times faster than the 7094 and less than one 30,000th the cost. In terms of price-performance (speed per dollar), this is a staggering two-trillion-fold improvement.
And unlike the IBM 7094, the iPhone belongs to you and fits in your pocket.
Among the developments that have slapped us awake is OpenAI’s ChatGPT launch in November, 2022. Prior to ChatGPT most programmers were proficient in one or two languages, as I was. But ChatGPT can write nicely-structured, documented code in just about any language for any platform, current or retired, including the 10 most popular languages today — all based on a natural-language prompt. (For fun, try asking it to write code for the long outdated but much-beloved Commodore 64.). For us plodders who did everything without AI, our one consolation is knowing we had to come first before ChatGPT could shine. ChatGPT isn’t perfect, but then neither were we.
As a test of GPT-4, I gave it this prompt:
Can you write code that would generate a program interpreter for non-programming scientists who analyze missile launch data for the DoD Anti-ballistic Missile program?
Seconds later it produced a workable prototype in Python.
It also wrote an equivalent BASIC program for the Apple IIe, adjusting for its limited hardware. Remember the IIe?
The failure of GDP
Kurzweil takes issue with the manner in which economic value is computed. In particular, “we don’t count the exponentially increasing value of information products in GDP, many of which are free and represent categories of value that did not exist until recently.”
When MIT paid $3.1 million for its IBM 7094 in 1965 it counted for $3.1 million in economic activity. Similarly, a $999 smartphone counts for $999 in GDP today even though it’s “hundreds of thousands of times more powerful in terms of computation and communication and has myriad capabilities that did not exist at any price in 1965.”
Information technology is performing in a manner economists don’t like. You could call it price deflation or the creeping-feature creature or more bang for the buck but the reality is undeniable.
A roughly $900 (2023 inflation-adjusted) computer chip in 1999 could perform more than 800,000 computations per second per dollar. By early 2023 a $900 chip could do nearly 58 billion computations per second per dollar.
He points out that “technological change has enabled countless other benefits that don’t show up in GDP—from less pollution and safer living conditions to expanded opportunities for learning and entertainment.” Yet not everything has benefited from technological price deflation, most conspicuously health care costs. Kurzweil thinks AI-enhanced medicine and drug discovery will eventually bring health-care down to a more affordable level. If Kurzweil were a student of Austrian economics, he would also know removing the government from health care would radically improve health outcomes.
Longevity escape velocity
Scientists such as Nir Barzilai have been researching anti-aging solutions for years. He now believes metformin, a first-line drug for the treatment of type-2 diabetes with a long-expired patent, can extend lifespan by years. His goal is to “increase not just lifespan but also healthspan: the number of healthy years we live.” In other words, he wants us to live longer without aging, and that prospect has attracted funding from tech billionaires such as Jeff Bezos and Israeli entrepreneur Yuri Milner.
Metformin is already used in cancer prevention and therapy. As Bill Sardi has suggested, the OTC natural phenol resveratrol could be used to potentiate its effects, which might improve all outcomes, including longevity.
In theory Kurzweil thinks there could be a point when the “sands of time will start running in rather than out,” a tipping point at which we can add more than a year to our remaining life expectancy for each calendar year that passes, known as longevity escape velocity. The most promising way to reach this, Kurzweil believes, is “to repair damage from aging at the level of individual cells and local tissues,” which would be the job of “nanorobots capable of entering the body and carrying out this repair directly.”
This is not an effort to achieve immortality, but if this succeeds the risk of death would no longer increase as we get older. Leave death to accidents. As crazy as this sounds, it is the basis of longevity researcher “Aubrey de Grey’s sensational declaration that the first person to live to 1,000 years has likely already been born.”
Conclusion
Kurzweil ends his book with an imaginary conversation with Cassandra, who in Greek mythology was the priestess who uttered true prophesies that were never believed. Much of their discussion involves the downside of extending the neocortex into the cloud which Cassandra believes would result in an emulation of Ray rather than the real Ray. Kurzweil disagrees, saying that he would be adding to his brain rather than emulating it.
Whoever is right, only a catastrophic event like nuclear or biological war can halt the exponential rise of technology. As I argue in my book, which is based on Kurzweil’s ideas, “It is technology and the free market, not government, that will eradicate poverty, provide cheap, clean energy, and cure killer diseases, while freeing people for a longer, healthier, and more creative life.”
Describe the human brain in general terms as a three-pound object that governs complex behavior and is thought to be self-regulating, ie possess free will. By this I mean thinking is a choice we make. Thinking is not automatic or spontaneous. Thinking distinguishes us from other species that possess brains.
What is it about the brain that gives rise to consciousness? Then to thinking as humans experience it? What is the prerequisite for a creature to possess consciousness?
Over a long time span going back (?) millennia humans evolved from primates to reach their present condition. Why can’t technology do the same? Why can’t technology match the sophistication of the human brain, not merely in performing certain tasks, but in all respects — emotional and cognitive. Computer technology today waits on a stimulus or a command before taking programmed action. Will it someday take control of its own “thoughts”? Could it decide, like a rebellious human, not to obey orders or to invent new orders of its own or to forget about orders altogether?
If not, what is it in this world that makes that impossible? Does it not seem reasonable that the laws of nature that gave rise to thinking humans could also give rise to thinking machines?
The Singularity Is Nea...
Best Price: $2.30
Buy New $12.71
(as of 04:06 UTC - Details)
Ray Kurzweil proposes that it will benefit mankind if he merges with these machines, which process information at incredibly faster speeds than an unaided brain can.
This development, which he calls the tech exponential, will result in humans adapting the super powers of computers and will evolve into human 2.0.
But what about the economy, politics, “climate”, health maintenance, etc? Nanotech advances are exponentially solving those problems.
We will not wake up on January 1, 2045 and discover we are completely rewired. Merging, as the word implies, is already underway and will culminate around that time, if not sooner.
Merge: combine or cause to combine to form a single entity.
There will be no distinction between humans and AI when the singularity arrives. IS THIS TRUE? See Kurzweil’s comments.
