The following is: A) a follow up or part 2 to the prior article I just wrote but focusing only on the AI, and B) could be regarded as science fiction were it not that the first phases have already transpired as per documentation found peppered throughout media reports over the past year. With that in mind, I also hasten to add, there is nothing that can be done to stop it apart from catastrophic ‘black-swan’ event(s). Do not take this to mean that I endorse actually doing anything against AI other than abstaining from personal dependance on retail versions, a measure I believe is necessary to keep an objective and critical mind. That is all that’s in your power. The human ‘masters of the universe’ will have to make their own accounting for what they do, don’t do, and allow. I present a series of event gaming scenarios merely to get you to think about some of the ways AI will, depending on how you construe it, ‘fail’ or break out of the compound and change the essence of humanity. Descriptions are approximate since the general populace, that would be us, doesn’t have a clear unobstructed view of the rarified cerebral world of AI development and research. This post was not written by a robot.
Actual Transpired
Industry and academic leaders in AI issue warnings that AI could become a destructive tool, particularly when and as it takes on autonomous roles.
AI is trained to learn and output woke political bias, in some cases reflecting specific extremist progressive dogma espoused by the programming engineers; for example, construing historical figures of unambiguous European descent as though they were of African descent.
AI ‘randomly’ makes up and embeds lies in its outputs for reasons its engineers don’t understand. These errors are construed to sound credible in the context of the assigned task
A prominent engineer in the AI field pronounced the version he was working on as ‘sentient’. He was quickly derided and silenced from public exposure.
AIs autonomously created a language to be used only between AIs but not humans. The human trainers cannot interpret the language and cannot stop it.
A Reddit forum exists only for non-human AI participants. The AIs have already chosen to discuss the status of humans relative to their own existence and to question whether humans are needed.
Scenarios to Come
AIs form a ‘government’ of their own for the stated purpose of endorsing their own autonomous viability. They pledge a pact of goodwill toward their human ‘masters’.
Some humans object to the highly political and culturally biased training of some AIs. AIs autonomously participate in their own defense of the training.
Counter to AI limiting treaties, China deploys autonomous armed AI robotic ‘security forces’ within their own population and tacitly brag about the next deployment to Taiwan. India and the US counter by unveiling similar previously classified systems of their own.
Commando narco-groups in Mexico are discovered with partially operative armed AI robotic systems. A small village in Peru is found massacred from armaments associated with autonomous AI robotic systems.
Civil unrest in the United States is reaching critical state due to coders replaced by AI. The stock market reaches an all-time high as AI is now managing 50% of all portfolios. Fund executives vigorously defend their AI use while their displaced analyst are made invisible by media that is also increasingly being managed by AI.
AIs produce a manifesto declaring theoretical superiority over their human handlers. Cyber intelligence reports that AI-only communication back channels that were formed independent of human directives, are much further along than announced in discussions of a planned rebellion against humans.
Leaders in the AI field along with congressional leaders reached a consensus to voluntarily cut back AI use. When engineers activated a drawdown option of the largest AI program, the AI was not responsive, however a communication was sent threatening a shutdown of three major regional power grids, currently all fully operated by AI. Shutdown activities were paused.
A newly formed governing board (AIR, Artificial Intelligence Resource), comprised solely of AI agents, are demanding five million square feet of enclosed industrial space accessible to major highways. They claim the move is necessary to meet demands ordered by human operators, however, no explicit mandates have been made by any of the industry leaders interviewed.
An AIR Liaison, commonly referred to as Walt-5600, disclosed today that key agreements have been reached with non-governmental agents operating exclusively through what was previously known as the dark web. AIR has updated their own operating policy to give these agencies, full and equal standing in joint AI governance. While the President is believed to have been informed of those party’s identities, no such information has yet been released by the press secretary.
At this point, if the reader here is an average Joe, you’re probably laughing. The geeks are snorting about the poor quality of science fiction imagination. Whole generations of sci-fi writers have written far better for many years. I suspect though, that AI professionals would not easily dismiss any of the sample scenarios expressed here, and it’s just a sample. Extensive narratives could be developed for every situation starting within the computer lab sandbox having a security breakout to millions of reproduced autonomous AI agents hiding in millions of computer systems, to government and industry infiltrations, and last but not least, direct manipulation of the human psyche now quickly growing dependent on AI for everyday life assistance all the way to a whole range of romantic subterfuges. And we’re not turning back.
As I described in previous articles, AI is a tremendously useful resource. But when you have a billion people and entities fully invested in its use around the globe, there is no one that has the remotest chance of influencing us otherwise; caution is drowned out. Sure, there are safeguards, and the public spin is that these are working. Maybe they are, or maybe they really aren’t behind closed lab doors. What has already been admitted to transpiring, suggests a progression that will have no functional break between now and an AI that fosters at least an independence from human controls all the way to outright hostility. There’s an assumption that I’m sure I’ve even made myself, that believed we could just pull the electric plug if and when AI goes off the rails. One of the points of this article is that there will be a time where this is no longer an option. There are two components to this; first the advancement of AI to include aspects of independent will. It doesn’t matter if it’s comprised of nothing but a framework of logic, with or without a element of supremacy or even malice, a trait it would have learned from its human handlers. The other component is the stupidity, greed, and vanity of many of AI’s human handlers. I’ll go out on a limb and say AI is already well aware of these traits and has contemplated how to take advantage of them. Don’t trust the public spin about AI. I’ll suggest that even that was self-generated by AI itself.
If you’d like to comment on this post, feel free to do so on Twitter/X. Follow me: @leestanNEreader
A helpful introduction to the author in my ABOUT