Michael Lewis, author of the remarkable 2014 book “Flash Boys,” recently wrote an excellent piece in Vanity Fair about the efforts to discredit him by some of the groups he exposed.
“Flash Boys” and his March 17 Vanity Fair print article are yet another chapter in a revelatory journey, as Lewis once again alerts people to the abusive patterns emerging from a combination of high-speed trading computers and under-appreciated Wall Street motivations, incentives and greed. He reveals how trust is monetized and abused.
“Flash Boys” is an amazingly detailed and clear articulation of how investors large and small are effectively “front-run” and fleeced by very rapid high-frequency trading.
It complements “The Big Short,” an equally enlightening book by Lewis about Wall Street’s rapacious pattern of disguising bad investments as good, using rating agencies and other trust-based mechanisms to trick investors. Indeed, it seems other peoples’ misguided trust is now the coin of the realm.
In “Flash Boys,” we are reminded that the tools of power take on many subtle forms, some too arcane and complicated to be recognized even by very intelligent and alert people.
“Flash Boys” implicitly raises a central question about the challenge of artificial intelligence (AI). Most important, AI is now capable of a stupefying level of computational speed and artificial complexity that no human can readily apprehend.
Moreover, AI is made most dangerous when it is both miscast and unappreciated. Its best disguise is our blank uncomprehending stare.
Physical computing devices make things happen, but it is the ethereal computer code that truly is the source of knowledge of good and evil. The code embeds and carries the necessary fragment of human nature for the specific tasks and objectives at hand, and thus it poses the real danger.
It’s not necessary for a thinking machine to spontaneously develop its own motivations. People themselves impose encoded motivations as part of the purpose to be achieved, whether for good or evil.
While it’s naïve to require prototypical artificial intelligence to take on an independent, aware and self-reliant capacity before it poses a threat, it is equally unnecessary for machines turn to attack and destroy their makers because of a native self-awareness before we block that behavior.
An AI machine that can politely censor speech is dangerous. Virtual reality is inherently not truthful. Machines programmed to arbitrage or take advantage of human motivations, fears, conflicts, needs or reasoning hold seeds that are not harmless; each AI-induced harm can be slight, chronic, parasitic, acute or lethal, and may shift from one to another in a millisecond.
Put another way, superhuman thinking machines with incomprehensible capabilities encoded with malevolent human purpose pose real dangers. Unfortunately, the main purpose of many physically endowed AI machines is to do evil destructive things without hesitation.
Elon Musk recently spoke of AI as “summoning the demon” and subsequently donated $10 million dollars to the Future of Life Institute (FLI) in an effort to keep AI from becoming evil.
In an open letter, the FLI says: “we cannot predict what we might achieve when this intelligence is magnified by the tools AI may provide,” and later, “we recommend expanded research aimed at ensuring that increasingly capable AI systems are robust and beneficial: our AI systems must do what we want them to do.”
While this is an admirable objective, humanity has no specific Ten Commandments for intelligent machines and thus seemingly lacks a sense of what is good and evil.
FLI’s sentiment also seems to sidestep currently disguised evils, the actual real-life examples now ongoing that are generating billions of dollars in profits to a tiny group of people who are exploiting a mere shred of what modern programmable computing capacity can accomplish.
In fact, high-frequency stock trading machines are doing exactly what that particular group wants of them, and therein lays the problem: Who is the favored “we” imagined in the FLI letter?
The greater danger to the majority of humanity rests in the mischievous motives of the relatively small group of empowered people — essentially tribes — who set these remarkable machines upon their specialized tasks. That is behavior legal systems have managed for centuries, but for which we now lack cogent enforcement when intermediated by AI machines.
That is precisely why the warnings of a benevolent genius like Michael Lewis must be understood and not merely ignored. His expositions are warning us all about the demon that is already among us acting in a substantially uncontrolled fashion, one that is gaining political benefactors through vast wealth and purchasing protection via focused lobbying and political contributions.
As the thinking criminal machine, Scarface-AI, says: “First you make the money. Then you get the power. Then you get the women. Ah, I’m AI; forget the women!”