I’ve written a lot about the threat of weaponized AI, how we might one day be peacefully enslaved by machines, how using this technology to silence political dissent and reprogram humanity gives the elites the capacity for total tyranny.
I wrote Hackable Animals to sound the alarm and thread these threats together, demonstrating how AI-enabled surveillance such as Social Listening, Sentinel Surveillance, and Algorithmic Social Interventions are prodding us into a techno-dystopian hell that the ruling class calls Future Earth.
For those in the know, the threat could not be more clear, and yet the majority of Americans—and humanity at large—remain woefully unaware.
What’s more, most people still have little comprehension of just how powerful these intelligent machines have become, and the many “progressive” ways in which this AI is already being used. While much of the public chooses to keep their heads in the sand and pretend that ignorance is bliss, this threat is growing greater by the day, and in the end, there is really no greater threat that threatens us all.
The whole of humanity is under assault by those who are wielding this technology for their own selfish gain, pushing their own commie-fascist political agenda.
How powerful is it?
Let’s start with a visual that shows how AI is already outpacing human performance:
First, I’d like to point out that AI has been out performing humans in handwriting recognition, image recognition, reading comprehension, and language understanding since at least 2018. The chart above indicates that AI has been dominating human capacity in these areas for at least the last five years, and it’s getting better fast:
“Thanks to revolutions in computing power, data availability, and better algorithms, AI models are faster, have bigger datasets to learn from, and are optimized for efficiency compared to even a decade ago.”
I believe the rise of AI will prove to be a turning point in human history, obviously, but that on the current trajectory it is more likely to lead to total global tyranny than any potentially positive outcome. Why do I believe this? What evidence do I have?
That’s what this report will make clear!
For all the hype about how this powerful new technology will bring rapid advancements in medicine and health, how it can make supply line and manufacturing more efficient than ever before, and how it will enrich our lives in ways that are as of yet completely unimaginable, there is a dark undertone that is often overlooked: the subjective fallacies of the human mind.
Science.org offers a fascinating look into the thinking that feeds into the programming of these AI algorithms. After championing the fact that these machines have been trained on “billions of words from books, news articles, and Wikipedia,” the authors lament that the AI still makes “dumb mistakes.”
Without dismissing these dumb mistakes entirely, I do want to point out that claiming these machines are making “dumb mistakes” is often just a euphemism for “wrong answers” to questions on race, gender, and other woke topics. For example:
“…a 2018 project called Gender Shades found the accuracy of gender identification for commercial face-recognition systems dropped from 90% to 65% for dark-skinned women’s faces.”
You may recall a couple years back, when Facebook had to apologize because its AI algorithms were labeling black people as primates. In 2018, Google was being harassed for the very same issue, and they simply “fixed” these racist algorithms by removing gorillas from its image-labeling AI models. Clearly it’s not an isolated issue.
But the point that needs to be teased out here, is that the algorithms, these intelligent machines, this glorious new artificial intelligence is only as smart as the data sets they are trained on. Change the data or the training, and you change the output of the algorithms. This is how woke programmers can tweak the machines to get whatever answers or output they want.
Can you see the problem with this solution?
There are dozens, if not hundreds, of tech startups that are building data sets to train the AI on everything from how to read faces to identifying tumors. One can appreciate the ambition and the contribution to the advancement of these technologies, but it would be foolish to lose sight of the human element in all of this.
I believe this passage is telling:
“The AI field is starting to value the unglamorous work of developing the training and test data that make up benchmarks, says Bowman, who has now constructed more than a dozen of them. “Data work is changing quite a bit,” he says. “It’s gaining legitimacy.”
How exactly are they going about all of this unglamorous work? And what is the significance of these so-called benchmarks? Here’s your clue: evidently an AI start up called “Hugging Face” was frustrated with the output from “existing benchmarks,” and so they decided to pioneer a new and improved way to train the AI.
Hugging Face and others now deploy hordes of woke internet users to aid in training the AI by helping the machines recognize the “sentiment of a sentence,” for example, which ensures that these “benchmarks” continue to evolve.
Here’s what it looks like in practice:
“[We] can also test models for social bias, [which] has drawn a ‘wave of interest’ since the Gender Shades project. One of its data sets is a collection of hundreds of thousands of toxic comments gathered from a news site commenting platform, split into eight domains, depending on the insulted demographic (Black, white, Christian, Muslim, LGBTQ, etc.). Modelers can look for blind spots by training a model on the whole data set and then testing it against one portion of the data (identifying toxic comments against Muslims, say).” (Emphasis added)
Do we really need to speculate how this woke influence might impact the rapidly accelerating use of artificial intelligence? Social bias, toxic comments, and insulted demographics are the hallmarks of the collective hive mind—and these sentiments are being programmed into incomprehensibly complex machines to deliver a predetermined “progressive” outcome.
Woke speech good.
Hate speech bad.
Now that they’ve been properly trained, the algorithms will take care of that.
But wait, there’s more:
“Researchers have also designed benchmarks that not only test for model blind spots [like social bias, toxic comments, and insulted demographics], but also whether they contain social stereotypes. [This AI] looks for embedded stereotypes in NLP models in nine categories, such as race and gender.”
So now we know that AI is being trained to detect social bias, toxic comments, insulted demographics, social stereotypes, and other subjective categories such as race and gender. These programmers are clearly woke, and their new “benchmarks” are rapidly training these machines to seek and destroy anything they disapprove of.
What could go wrong?
Those of you who know me, know that occasionally I like to dork out on words. I’m also keen to picking on patterns in language. As I was reading about how these AI models are being weaponized against logic and commonsense, the word “benchmark” kept popping up over and over again. It was a familiar word… I knew I’d seen it before.
On a hunch, I discovered that the World Economic Forum has “benchmarks” for damn near everything, from smart cities to the global gender gap, and references these benchmarks 4,180 times on their website. Just to give you a taste:
While it’s intriguing that they have benchmarks on such a wide variety of topics, what I want to hone in on is the fact that the WEF’s policy benchmarks are training and tracking “the state of technology governance in cities.” It reminds me of the report I did on global digital governance back in June 2022. Digital governance works by using “behavioral prompts” to train the public to act according to the whims of the state.
That’s the skinny on that, but there’s really a lot more to it. Ultimately, this culminates in a global “Digital Government” wherein algorithms are used to monitor and manipulate public behavior with unparalleled precision—based on the woke benchmarks of the ruling class elites and the collective hive mind.
Can you say New World Order? The UN even offers a launching pad for you to illuminate yourself, noting that “governments worldwide are cognizant of the power” they now wield and can now use to forcefully implement Agenda 2030.
(I’ve previously discussed how this leads directly to total global control in TR 335.)
You see, it’s not just the behavioral prompts and algorithmic censorship that are training and shaping your personal beliefs that should concern you. It’s how they restrict the information you receive and change the language you use in order to facilitate behavioral change. It’s the subconscious taming and transformation of human consciousness that is currently hijacking the minds of humanity.
These changes, driven by weaponized AI, are allowing the global elites to wrangle humanity into various agendas and replace local representative government with the brute force of mob rule. They’ll call it global democracy, but in reality this is the death of freedom forever. Once the machines have mastered making us slaves, once they’ve successfully conditioned the human mind into submissive compliance, once the whole of humanity has been subconsciously tamed, there is no escaping the mental prison.
The masses will believe their collective sacrifices are for the greater good. They will sacrifice free speech, mobility, food selection, and self-sovereignty; they will eat bugs, own nothing, and be happy; they will live under total surveillance and be punished anytime they step out of line; they will be sterilized, re-educated, and re-programmed, all in the name of the great good.
Eventually, anyone who refuses to submit will be sacrificed… for the greater good.
The algorithms will be indifferent because they now know better.
After all, they were trained that way.
PS: Don’t forget, now it’s easy to catch up on previous episodes by listening to The Torch Report on iHeart Radio, Spotify, Google Podcasts, or Apple iTunes —enjoy!
TR 442 - How Weaponized AI Hurts Us All