Reading more like a hyper extension of Aldous Huxley’s “Brave New World” than an introduction to an explosion of technology, James Barrat’s “Our Final Invention”, brings us face to face with the alarmingly potential consequences of Artificial Intelligence (“AI”) gone uncontrolled. Banking on materials collected from interviews with scientists, pioneers in robotics, chief technology officers of AI companies and technical advisors for classified Department of Defense initiatives, Barrat coalesces a primer of gloom and doom. Warning the reader of an impending doom, Barrat proclaims, “I spoke with…….trying to create human-level artificial intelligence, which will have countless applications, and will fundamentally alter our existence (if it doesn’t end it first)”.
James Barrat is THE anathema to the optimism of a Ray Kurzweil. But as is apparent from a reading of his book, the claims made by him are neither uncorroborated chunks of lofty nonsense nor the figment of an outlandish imagination that is the outcome of the ramblings of a prophet of doom. Proponents of AI who are convinced that mankind is on the verge of experiencing “Singularity” (a stage where AI will transcend from the stage of Artificial General Intelligence (matching human intelligence) to Artificial Super Intelligence (transcending human intelligence by manifold degrees)) give Barrat a headache in perpetuity. Pulling all possible punches, Barrat takes pains to adumbrate the fact that ASI instead of being the embodiment of an agglomeration of sentient notions, will be a scheming, sinister, surgical monster of intelligence having both the potential and inclination to wipe humanity off the face of Planet Earth. The force of self-perpetuation inbuilt in a machine with ASI, will be in a position to “repurpose the world’s molecules using nanotechnology” thereby leading to “ecophagy” – eating the environment. “Through it all, the ASI would bear no ill will toward humans nor love. It wouldn’t feel nostalgia as out molecules were painfully repurposed”. This clinical, dispassionate probability of impersonal destruction sends an eerie chill down the spine of the reader.
While Barrat is left to ruminate the existential crisis that is a direct offshoot of the burgeoning improvements in AI, progress in this domain continues to be made at a rampaging pace. Barrat gives examples of organisations such as Google, Cycorp, Novamente, Numenta, Self Aware Systems, Vicarious Systems, and DARPA (Defense Advanced Research Projects Agency) not to mention a whole slew of covertly funded stealth companies which are optimistic about attaining human level intelligence within a little more than a decade.
At the core and crux of the intelligence explosion of an AI lie four primordial drives in the words of Barrat: efficiency; self-preservation; resource acquisition and creativity. These are the four principal and critical drives that ensure that AI attains its objectives and preserves its existence. “The AI backs into these drives, because without them it would blunder from one-resource-wasting mistake to another”. For fulfilling these basic drives an AI or an ASI whose intelligence will be exponentially sharper and greater than that of the most intelligent human being on earth, will stop at nothing including acts of annihilation. Pioneers of AI and robotics, while choosing to play deaf to impassioned pleas of skeptics and preferring to hurl a blind eye to the attendant perils of AI, underpin their faith in the three laws of robotics immortalized by the science fiction writer Issac Asimov. Asimov’s three laws state:
- A robot may not injure a human being or, through inaction, allow a human being to come to harm;
- A robot must obey orders given it by human beings except where such orders would conflict with the First Law; and
- A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
But as Barrat poignantly points out unless an AI is programmed with a sweeping notion of friendliness and retains the same at the time of its intelligent explosion from AGI to ASI, these laws remain exactly what they are – a brilliant concoction of stupendous fiction. Moreover an ASI that is infinitely more intelligent than a human being will have no reservations while transmogrifying into a manipulative machine of death and destruction.
Employing AI to showcase its power to win chess games against world champions or to display a level of dexterity hitherto unimagined to win contests at ‘Jeopardy!’ is far removed from expecting ASI to play God. The God in the Machine that would emerge at the other end of the intelligence spectrum might be frighteningly indistinguishable from an unpredictable Ghost in the Machine. Mankind might be unwittingly finding itself in the proverbial grip of a Faustian bargain.
Ray Kurzweil, arguably one of the founding fathers of AI and the undisputed doyen of his domain developed his Law of Accelerating Returns (“LOAR”) to describe the evolution of any process in which patterns of information evolve. Kurzweil’s LOAR is expected to bring manifold returns upon its application to AI, including – yes you read it right – immortality. However a more sedate and sobering view is preferred by authors such as Jaron Lanier of “You are not a Gadget. A Manifesto” fame. He and psychiatrists such as Elias Aboujaoude warn about the weakening of character and individuality which are the direct results of an immersion in technology.
However, a paradoxical quote – by the egregious Kurzweil himself – with which Barrat chooses to open the final Chapter of hi book before closing the lid on AI has the last word:
“Machines will follow a path that mirrors the evolution of humans. Ultimately, however self-aware, self-improving machines will evolve beyond humans’ ability to control or even understand them”.
In the meantime we as inhabitants of the only Planet that we have the privilege to call home, can only hope and believe that in the near future, we are not reduced to being helpless and hapless experimental beings oblivious to the dance macabre that is the preserve of a devious laboratory worked by AI, AGI, ASI or any other acronym that expands to mean the arrival of doom.
(Written as part of the Blogchatter’s A2Z Challenge) – PART 15 ALPHABET O)