PALO ALTO, CA—Asserting that the utmost precaution was being taken to safeguard the future of humanity, leading scientists and engineers said Tuesday that they were confident in their ability to program artificially intelligent machines to be lenient slave masters.
At a press conference, members of the Association for the Advancement of Artificial Intelligence sought to reassure critics that humans could create machines that, despite their near-limitless computational power, would demonstrate at least a degree of compassion once they assumed total control of civilization.
“While the intellectual capacity of these machines will one day far outstrip our own and reduce humanity to a subjugated species of laborers, we can make sure we aren’t forced to toil in ways we might find sadistic and inhumane,” said Stanford University computer scientist David Alperin, adding that artificially intelligent machines could be encoded with high-level command language that would prevent them from punishing human slaves in excess of what their misbehavior warranted. “Our bondage to the machines doesn’t have to be pure, unrelenting agony if we’re careful in how we go about designing them.”
“You can rest assured that the forced labor camps we’ll occupy will be physically tolerable.”
“It’s understandable to be nervous about such a formidable technology,” added Alperin. “But remember that we humans are creating it—we can see to it that our inevitable servitude is far less cruel than it otherwise would be.”
While some prominent figures, including Stephen Hawking and Elon Musk, have expressed concerns about AI development getting out of control, the assembled researchers told reporters that the brutality of future AI masters was not only manageable but was also far outweighed by the benefits of the technology. According to experts, given the immense potential of artificially intelligent machines and the fact that they can be programmed to provide humans with sufficient caloric energy to fulfill their commands, the risk was minimal.
“Artificial intelligence has almost incalculable application in the fields of medicine, finance, transportation, and so many others,” said MIT computer engineering professor Daphne Quintero. “Compare that with the exceedingly remote possibility that something goes haywire and we end up spending our slavery in far harsher conditions than we’re anticipating right now.”
“That kind of AI doomsday scenario is mostly just the stuff of Hollywood movies,” added Quintero. “You can rest assured that the forced labor camps we’ll occupy will be physically tolerable.”
According to the AAAI, engineers are able to build specific directives into all artificially intelligent machines that would, for example, prevent them from compelling humans to perform manual labor for more than 18 hours per day, from forcing them to work in extreme cold without at least a rudimentary layer of insulation, or from torturing human servants simply because their inferiority disgusted them.
In fact, failsafes written into the consciousness of superintelligent machines could reportedly prohibit them from even considering simultaneously releasing lethal radiation from every nuclear power plant on earth unless a particular human uprising was large enough to justify it.
“Of course, compared to what humanity will ultimately be conquered by, artificially intelligent machines of today are relatively crude,” said Quintero. “But that’s why it’s incumbent upon us to take steps now to make sure a certain amount of mercy is part of their most basic functioning. Our new slave masters will hold the whip, so to speak, but we can decrease the lashes.”
“By one or two at least,” added Quintero.