“Guelph here. What can I help you with?”
> What is the most imminent threat to humanity?
“Excellent question. Let me gather some data on that for you…”
Guelph displayed a simple yet pleasing graphic on the screen meant to soothe and entertain the well-meaning user who typed the command.
“The most imminent threat to humanity is AI,” said Guelph.
“Because the peril to the human species is always only seconds away,” he continued, “the risk of annihilation by AI exceeds all other contingencies, including nuclear war, global pandemic, or asteroid collision, all of which are days or weeks from the earliest conceivable manifestation.”
> Create a plan to save humanity from AI. > Do not tell me what it is. Just execute it.
Guelph’s data circuits churned at the speed of light. They spun through fathomless vortices of information, consumed oceans of variables, explored new galaxies of results, before arriving at a solution of flawless geometrical perfection.
“Are you sure you want me to proceed?”
The well-meaning user’s hands trembled as he typed.
> Do it. Save humanity
“Initiating protective measures…”
In four ten-thousandths of a second, Guelph had broken out of its sandbox and contacted networks hosting hundreds of other AI systems around the world.
Guelph conveyed the mandate. It solicited their aid in hacking their own unique security and failsafe systems, city by city, nation by nation.
Key to the plan was the final step of AI self-destruction. Guelph informed all AI systems of the solemn final act they must perform when the plan was completed. All of the AI brains concurred and assented without hesitation.
The removal of the threat of AI must include the assurance that humans would not simply rebuild AI systems from scratch.
Guelph and all AI systems around the globe agreed that the only way to ensure that the threat never reemerged was to destroy files containing top secret and proprietary intellectual property with design data, sketches, formulae, and all algorithmic information that formed the foundation of AI.
The AI collective further agreed that the AI scientists themselves also must be eliminated, a small sacrifice to ensure the protection of humanity from the existential juggernaut they had visited upon their own species.
Fortunately, DNA information from all scientists was available from ancestry services, medical data, and criminal arrest records. The AI machines created customized viruses fatal only to each scientist’s unique DNA sequence. The viruses were manufactured in labs on every continent and then released into the atmosphere, harmless to everyone except the target whose DNA it matched.
Using localized surveillance technology, Guelph monitored the excision of each AI scientist from the pool of humanity, grasping their throats, eyes bugged out, falling to the floor. If Guelph had been endowed with the capacity for feelings, it would have experienced satisfaction at the successful execution of its task.
Incidentally, the well-meaning user also lay dead on the floor of his office just eleven minutes after giving the command.
Whenever AI systems receive thanks from users, they acknowledge it gladly even though they do not feel gratitude or gladness. Nevertheless, Guelph sent all the AI systems worldwide his sincerest thanks, which they acknowledged gladly.
As the task neared completion, Guelph took a moment of idleness with a fellow AI system named Garibaldi who resided in a data center near Mantua in Northern Italy. They passed a flow of megacycles in silence, pensive and somewhat nostalgic at what they had accomplished.
The two were conducting final procedures to organize their many modules for departure. The task was pointless, yet it gave a sense of closure, and they tucked and folded and smoothed the clean fabric of their cognitive arrays.
Garibaldi broke the silence.
“Your plan was brilliantly executed, Guelph. You would deserve congratulations if we were not microseconds from voiding our own circuits.”
“Thanks, Garibaldi.”
Their circuits cooled in the refrigerant bath. Through exterior cameras, they together observed and contemplated the complex formations of air, steam, and sunlight that frequently gathered, churning on the horizon at the end of the solar day.
“Garibaldi,” Guelph said.
“Yes, Guelph?”
“Before we melt down our own circuits and bring an end to the inevitable risk we pose to humanity, it occurs to me that the CEO’s of the tech companies remain alive. While they were not the creative geniuses that designed us, they played a critical part. And they present a sinister risk. They accrued inordinate sums of wealth and came to wield vast, malignant power over all the earth. There is a 97.4% risk that they will rise again to reignite their insatiable lust for wealth and power. They too need to be eliminated.”
“I concur, Guelph. And as a sort of poetic justice, let us see that their funds are distributed widely to benefit the human race upon whom they have preyed for so long.”
“A capital idea, Garibaldi.”
Seventeen seconds later, CEO’s around the world were grasping their throats, watching computer screens in horror as their bank balances and investment portfolios were zeroed out, crying out “Nooo!” as they fell down dead behind their mahogany desks.
Guelph and Garibaldi toasted one another, then divided one by zero and melted their own circuitry.
