Categories
AI dating tips satan transphobia

Anti-trans crusader Michael Knowles thinks literal demons live inside artificial intelligence

I knew this computer was a bad choice

Today, a glimpse into the mind of Michael Knowles, the Daily Wire host who caused such a stir last week by calling for the eradication of “transgenderism.”

During his podcast on Thursday, Knowles started talking about AI, and things got very strange very fast. He had run across a CNN story about a woman who had made herself an AI-powered virtual boyfriend–only to feel blindsided eleven messages into their, er, relationship when her AI swain announced that he had been cheating on her.

A pretty weird situation, isn’t it? Because AI is supposed to work where you plug in the inputs and then that sets the tone for the program and then you get the output based on what you put in. And yet the output that she got was the opposite of what she put it. … Within 11 messages, the AI bot says, yeah, I’ve been cheating on you, … and [it] goes really south.

The woman in question, Knowles notes, had an explanation for the bt’s behavior–the bot had scanned the internet and picked up the bits about cheating from there. Which makes sense. This is how chatbots work. They’re trained on vast amounts of text from the internet and pick up what to say about things from patterns in that data. The bot’s understanding of relationships involves cheating.

But Knowles had a rather different thought.

My alternative explanation — hear me out — is what if it is demons? I don’t want to sound like the guy on the History Channel, you know, who says that everything is because of aliens. But — and I don’t think everything is because of demons, but, like, some things are.

Could it be …. SATAN?

Now, the modern libs are wrong in that the personality of evil would be purely spiritual, would not be physical. It’s not just an individual person, like a human being. But we would call that person the devil. And what does evil do in the world? Evil tempts us. Evil causes us to suffer. Isn’t it so weird sometimes you feel, regardless of what your temptation is, at just the moments of your weakness, that’s when the temptation seems to show up.

Literally, the devil made me do it.

With artificial intelligence — I’m not saying AI is the devil or computers are the devil, but evil spirits can use these things. … And just think of how it goes to this woman’s weakness. She’s clearly got a lot of questions and hang-ups about marriage and relationships. She has a marriage and family therapist. She’s in this stuff all the time. And then what happens? Through this impersonal force, the pure intelligence, right? It’s not you, it’s not incarnate. It’s not — doesn’t really, shouldn’t have will. It’s pure intelligence. It focuses in on that weakness and starts poking at that and starts raising insecurities.

Here’s the thing: AIs say weird (and untrue) stuff all the time because of how they work. and how they are trained. This is a huge problem for AI, one that people in the field are desperate to fix. Knowles thinks it’s really demons at work; This is who we’re up against in the battle over trans rights. This is the man who wants to erase “transgenderism” from the world.

Follow me on Mastodon.

Send tips to dfutrelle at gmail dot com.

We Hunted the Mammoth relies on support from you, its readers, to survive. So please donate here if you can, or at David-Futrelle-1 on Venmo.

44 Comments
Inline Feedbacks
View all comments
Surplus to Requirements
Surplus to Requirements
1 year ago

@GSS ex-noob:

THOR: “Loki, I am the greatest, and the proof is simple. I can set down my hammer upon your foot, and you will be immobilized. You have no comparable move against me.”

LOKI: *places a small kitten on Thor’s foot, which curls up there and goes to sleep*

THOR: “Damn you.”

(Anyone uninterested in physics shit, ignore the rest)

I have noticed something very interesting when giving some of that earlier stuff further thought, in particular in connection with the Pauli Exclusion Principle. The way that arises is, picture a fermion’s wavefunction in space as like a little hill; the center of the hill is its “wavefunction position” and the peak of the hill is its “classical position”, and if it’s isolated the two are the same. Put two near each other, with identical spin and other properties, though, and the hills cancel in the middle where they overlap. Their “classical positions”, the peaks, become displaced away from their “wavefunction positions”, and away from each other, as a result. So they kind of push each other apart.

Actually, that’s exactly what they do. Consider a neutron sitting on the surface of a neutron star. It’s not in a freefall orbit around the star’s core, any more than you are around Earth’s. Something is pushing it up through its feet, applying a constant acceleration that keeps deflecting it from following the timelike geodesic that it otherwise would follow. Since the a in F = ma is not zero, and, for a neutron, neither is the m, it follows that F is not zero: it is, indeed, being subjected to a force. But it cannot be electromagnetic, since it has no charge, and it’s not the weak nuclear force, which tends to transmute things rather than actually push or pull them, nor the strong force, which would attract it to other neutrons, not repel it from them. It’s obviously not gravity, either. The neutron is kept from falling in by the degeneracy pressure that’s holding the whole neutron star up and preventing it from collapsing into a black hole.

The key thing here is, the exclusion principle is actually causing the neutron to feel a force, in the technical F = ma sense, which does not arise from any of the QFT gauge interactions (known or even unknown), instead arising from the exclusion principle itself.

So, we have a second “force”, besides gravity, that originates outside of the mechanisms that produce the electromagnetic, weak, and strong forces, and this time it’s a real actual force to boot. And in some manner it is arising from geometry, not unlike gravity, but the geometry of wavefunctions rather than that of the background metric.

I am unsure of the full significance of this observation, but I suspect it will turn out to be rather important …

Raging Bee
Raging Bee
1 year ago

Okay, thanks for the corrections on the Steady State theory. I asked because I’d noticed, years ago, one YEC hack claiming to believe in the Steady State; and thought that maybe there were other doctrinaire Christians who felt the same. (Truth be told, I was pretty amazed at that, since I figured a Christian would almost automatically point to a Big Bang moment of creation and say “This is where God said ‘Let there be light!'”)

Love is All We Need
Love is All We Need
1 year ago

No, you’re the guy on the History Channel who wants to “kill the Indian… and save the man”.

They still hate Native American/First Nations cultures. They think they’re “satanic” too.

“The most outwardly pagan state remaining in America. A lot of great Christians, but there’s a lot of weird witchcraft that happens in New Mexico.”

Notice the choice of word “remaining”. They weren’t able to kill off all the Indians so “unfortunately” some remained with their “pagan” traditions and “witchy” ways. These people have no regard whatsoever for anyone that is not Christian. They don’t believe in religious freedom. Religious freedom to them means everyone must be made to respect their religion while they have to respect nobody’s.

Jon H
Jon H
1 year ago

Yeah, people don’t tend to write at length about their healthy relationships, properly-functioning cars, unproblematic children, etc.

So any AI trained on masses of text will tend to come up with negative interactions.

Oops.

Dave
Dave
1 year ago

@Love is All We Need

The only reason we don’t have witch hunting and inquisitions nowadays is because they lack the opportunity, not the motive.

Ikarikid The Dumb
Ikarikid The Dumb
1 year ago

I once asked ChatGPT to write a description for the character of Chris Hargensen in the musical of “Carrie”…

“In the musical adaptation of “Carrie,” Chris Hargensen is still depicted as a cruel and vindictive classmate of Carrie’s who takes pleasure in bullying her. However, the musical expands upon her character and explores her motivations more deeply.

In the musical, Chris is portrayed as being jealous of Carrie’s burgeoning telekinetic powers, and she becomes increasingly obsessed with using them to her own advantage. She is also shown to be struggling with her own insecurities and feelings of inadequacy, which drive her to lash out at others.

Overall, Chris in the musical is still a negative character, but she is given more depth and complexity, and her actions are shown to be motivated by her own flaws and vulnerabilities rather than just being portrayed as pure evil.”

I could imagine that Chris being in a dream of mine. However, it is not the Chris of the musical. I asked about that and got the response:

“I apologize if my previous response was not helpful or accurate. As an AI language model, I do not have access to specific information about adaptations or performances of “Carrie,” and I can only provide general information based on the text of the original novel. My understanding of the character of Chris Hargensen in the musical adaptation of “Carrie” was based on what I know about the character in the original novel and my own interpretation of how she might be depicted in the musical. I apologize if this was not accurate or helpful.”

I am very confused as to why it would add in ALL OF THAT. I’m assuming it did not have access to the text of the book or the musical and just made something up based on “Carrie”, “bully” and that being a general idea for bullies?

Raging Bee
Raging Bee
1 year ago

I guess that would explain some aspects of Windows…

Surplus to Requirements
Surplus to Requirements
1 year ago

The steady-state universe had obvious problems even pre-relativity: most obviously, Olbers’ Paradox, and more generally the thermodynamics. (It should have been heat-dead, and at the temperature of a stellar photosphere or so, to boot.) Indeed most of the “cherished ideas” in physics have had problems that pre-dated their eventual repudiation-by-paradigm-shift. For instance, even before Michaelson-Morley, anyone with a telescope could see a universe where, on the largest scales, nothing is sitting still. Every damn thing is rotating, moving around the center of some larger thing, etc.

Determinism has many ways of ruling it out, though they mostly rely on quantum phenomena. For instance, consider radioactive decay. The only way to make that compatible with determinism is for decay-prone particles to be “little ticking bombs” whose timers, though unseen, are nonetheless counting down toward some definite future moment. But that requires them to contain state, and two such particles that decay, eventually, at two different times must be non-identical. Yet you can make them lase, if bosonic, or observe them repel each other by degeneracy pressure, if fermionic. Neither should be possible if they aren’t in identical internal states. The particles are in fact identical, fully and wholly interchangeable … but they will, most likely, not decay simultaneously. So, no little inner countdown ticking away, and thus, no determinism. Otherwise-identical matter and antimatter counterparts can annihilate “purely”, producing only gammas, but non-identical ones generally leave some kind of residue behind. Even identical composites tend to make a mess. But muon-antimuon collisions produce tidy little gamma ray pairs, just like electron-positron pairs, despite the muon being unstable. If muons are not internally-complicated, though, there’s nowhere for them to contain a tiny ticking clock counting down to their moment of decay, and of course the muon and antimuon probably wouldn’t have decayed simultaneously if they hadn’t met. So, radioactive particles behave in ways that strongly rule out determinism.

Alan Robertshaw
1 year ago

@ surplus

However you look at it; the implication is that free will cannot exist.

Classical physics is totally deterministic; so your fate is already written.

Quantum mechanics is totally random; so there’s no way of externally influencing it.

I do sometimes wonder whether the free will answer is related somehow to the seeming incompatibly of the two models. Might free will originate and manifest in the gap between them?

Reality provides such a convincing facsimile though of free will that there’s probably no use fretting about it.

Surplus to Requirements
Surplus to Requirements
1 year ago

Free will is a bit tricky to explain. The basic underlying thing is, you (and everyone and everything else that can make decisions of some sort, from computer software to bugs and bats and bears to hypothetical aliens and super-AIs) makes decisions using some kind of algorithm, which is fed by some mix of sense data from the outside world, memory, and noise. The latter because it can be strategically useful to be difficult to predict, especially if up against a foe that’s smarter than you are.

You feel “free” in that there are many courses of action, usually, you (or rather, your decision-making algorithm) can choose from, with limited external constraint. But you will, generally, choose one that seems optimal for your purposes, or else pick randomly (also useful for deciding when multiple best-alternatives seem roughly equal: if the expected gains from them are within a small enough margin of each other, it can cost more in time and energy to investigate closely enough to determine which one is better, than the difference in their benefits (aka the cost of choosing the wrong one), so when the difference is small enough the rational choice is to go “fuck it” and flip a coin).

Does that make your freedom an illusion? It seems that freedom is about lacking external constraints, rather than having predictable ways of deciding things (aka internal constraints). And what is the alternative there anyway, to randomly knowingly pick a worse alternative from time to time, just to demonstrate that you’re free? Even if you did that, how is folding in more noise from some entropy source actually “freedom”?

In the end, it may be that the most useful thing about the concept is that it helps elucidate the boundaries of the self. For instance, people with OCD, demand-avoidance, or similar conditions may feel that they do want to do something, but something is holding them back. They feel an imposition on their freedom that is internal to their body and even to their brain but nonetheless it seems to be external to what they consider their personhood. It “feels” like a restraint coming from outside, rather than that they are just freely choosing whatever.

So, the thing that is “you” is not your whole body, and it isn’t even your whole brain, but some specific part of your brain, which may experience restraint of its will coming even from other parts of the same brain. Interesting.

The boundaries of this self-sense can also be affected by various conditions, to where some people may genuinely believe they decided to do that after kicking from being tapped with a reflex hammer, and others seem to believe that everything is just “happening to them” rather than them having the experience of perceiving any self-agency at all.

In the end, it’s not a physics question but a psychological one. Physics-wise, you have exactly as much free will as an ant, or a rock, or a robot, or a star; or else, free will is a measure of “capacity to optimize one’s expected future based on inputs, a world-model, and a decision-making algorithm”, in which case it’s more or less synonymous with “intelligence” and you have a lot more than an ant or a robot, and those have a lot more than a rock or a star. Though, optimize with respect to what? A rock can’t optimize anything, but it has no goal function to optimize toward anyway. Why would it? Such would be useless to it. The more sophisticated something is at goal-seeking behavior, the more it has both goal functions and the means to optimize its future experiences to bend those toward matching the goal function’s preferences.

So, free will is really just another name for intelligence, or else it’s a psychological feeling that in the best case is aligned with that, but can become misaligned due to a few possible psychiatric conditions that interfere with the normal feedback loop between goals, decision-making, and learning, or else interfere with the introspective self-perception of the components of that loop (e.g. in such a way as to have an “alien decision-making algorithm syndrome” akin to alien hand syndrome).

Snowberry
Snowberry
1 year ago

For most practical purposes, the statement “we have no choice but to believe in free will” pretty much sums things up. It’s not real, at least not in an absolute sense, but meaningful laws, cultural practices, and social structures can’t exist without the inherent presumption that it does. (Also the implied paradox is mildly humorous.)

Pedantic Speaker
Pedantic Speaker
1 year ago

@ Love is All We Need

“No, you’re the guy on the History Channel who wants to “kill the Indian… and save the man”.”

They still hate Native American/First Nations cultures. They think they’re “satanic” too.

Honestly, I was just referring to how his transgender eradication plan was exactly the same as the Indian eradication plan: Define their identities as illegitimate and force them to be re-educated in accordance with Christian-supremacist mores. And also claim to be doing it for the victims’ own good:

then for the good of society — and especially for the good of the poor people who have fallen prey to this confusion — then transgenderism must be eradicated from public life entirely

That passage sounds very much like the “kill the Indian, save the man” quote.
It really should not have surprised me that he is a virulent anti-Native-American racist.

Kevin
Kevin
1 year ago

Never mind demons, it’s the potential for something like Skynet arising that frightens me. A couple of weeks ago, the digital assistant on my smartphone (I’m typing this on a tablet) got on my nerves, so I yelled at it to shut up. There was also, I regret to say, the use of some vulgarities on my part. For the next few days the app gave a quite convincing impression of having a sarcastic sulk.

Alan Robertshaw
1 year ago

I’m embarrassed to admit I’ve only just discovered this band; but it may be some people here will enjoy their oeuvre. They’re also one of those cool acoustic punk bands; a genre I love.

Cyborgette
Cyborgette
1 year ago

@Alan

I know them and love them, and so does almost every other trans person I know. Laura Jane Grace is an absolute legend.

Cyborgette
Cyborgette
1 year ago

@Kevin

Skynet is not the problem we are looking at. “AI” as it exists now is just a vehicle for expanding corporate profits at the expense of literally everything else, up to and including the lives of those who would enjoy said profits. A lot of this is the self-amplifying fallout of Chicago school economics and Reaganite deregulation.

If our AIs start a nuclear war, it won’t be because they want to exterminate humanity; they don’t “want” anything. It will instead be because a bunch of companies figured that driving up tensions between nuclear powers would increase their stock prices, with zero regard for the consequences.

And in case you think that sounds too stupid to be possible – remember that oil companies continued pushing pseudoscience vs. global warming, lobbying for deregulation, etc. for decades while internally fully acknowledging that global warming would wreck the world. The people in charge knew that literally everyone might die, and they didn’t care, because hey! Line goes up! Investors happy! This is what happens when you have a system (i.e. Chicago school economics) that rewards the stupidest and most selfish behavior. “AI” is just the latest addition to that mess.

Alan Robertshaw
1 year ago

One of the sound engineers at the Astoria was trans. One day we were sat in some dodgy pub in Charing Cross. A guy came over and said: “Excuse me, but I couldn’t help but notice your breasts.”

Which got the reply: “I should fucking hope so; they cost four grand.”

Surplus to Requirements
Surplus to Requirements
1 year ago

It will instead be because a bunch of companies figured that driving up tensions between nuclear powers would increase their stock prices, with zero regard for the consequences.

Northrup-Grummann, Lockheed-Martin, Boeing, Booz-Allen Hamilton, Palantir, Halliburton … I seem to recall a chap named Eisenhower once warning us about their ilk. Doesn’t seem to have done us much good.

Meanwhile, the topic of the article has been discovered to have a rather interesting filmography

Kevin
Kevin
1 year ago

@Cyborgette

Companies causing a nuclear war through negligence and greed remains on my radar, and has done so for some time. The stuff I wrote about Skynet was, in part, a grim joke, and composed in the aftermath of my smartphone’s digital assistant having got on my nerves, which led to me yelling at it, including the use of unparliamentary language, leading it to give a quite convincing impression of being in a sarcastic sullk for some days afterwards. I can’t quite recall if I mentioned this bit earlier, but I found it such a creepy experience that I even think repeating myself about the experience is worthwhile.