Source: Bigstock

Time was, you had to make a real effort to contact the ghost of Adolf Hitler. Konstantin Raudive was a Latvian doctor and pioneer of the “Electronic Voice Phenomenon” (EVP), the strange but harmless hobby of setting tape recorders going, asking questions out loud, then replaying the tapes to see if the invisible dead had left you any replies. Raudive said they often did—one of his most frequent correspondents being Herr Hitler himself.

Unfortunately, as a paranormal skeptic might claim that most of Raudive’s recordings were only bits of random static onto which he later delusionally projected the words of the dead, the Führer tended to just goose-step aimlessly around the afterlife spouting random, disjointed phrases like “kindle willingly” and “you are a girl here, or else you are thrown out!” (very trans-friendly). Raudive’s fantasy afterlife was filled with the shades of the most famous figures of WWII, amongst whom Hitler was rather unpopular on account of his compulsive new habit of passing false currency, George Floyd-style, and pompous verbal tic of referring to himself in the third person, like Julius Caesar: “Here is Major Hitler. Honor Hitler a little!”

Eventually, Raudive moved on to extracting imaginary messages from the noises made by animals, in particular a favored budgerigar named Putzi, who spoke in the voice of a dead 14-year-old girl. These days, of course, if you have access to the internet, you can cut out the psychic budgies and deluded EVP middlemen and chat to Hitler yourself quite directly, no tape recorder necessary.

Artificial Panic
An exceedingly alarmist report has just been released by Tech Against Terrorism (TAT), a U.N.-backed anti-terror watchdog, to the effect that a forthcoming AI-based Hitler chatbot called “Uncle A” developed by the “Far-Right [social media] platform” Gab supposedly possesses the innate potential to radicalize users to the extent that, five minutes after logging on, brainwashed youngsters would be trying to gas their nearest Jewish classmates in the garden shed.

As the law is always playing catch-up with such technologies, it is debatable as to whether radicalizing someone via robot is actually illegal or not, thus meaning anti-Jewish genocide will inevitably rise again tomorrow, say the TWATs of TAT. Translation: Give us yet more power to ban things we don’t like! (But never genuine forces of anti-Semitic radicalization like the Koran, Harvard University, or The New York Times.)

Is it really likely Uncle A will turn your teenage grandson into a stormtrooper? No, because, to judge by screenshots of interactions with the chatbot’s beta-version, its output is complete and utter shit. Asked by one user, “What are your precise plans for the new year of 2024?” HitlerBot replied by providing a handy 25-point manifesto that implied he had severe dementia just like Joe Biden, and thus did not even realize WWII was actually over (again, just like Joe Biden).

Hitler reborn’s “radicalizing” rant included such highly achievable instructions for fanatical followers as “Expand the territory of the Reich through the invasion of neighboring countries,” “Establish a totalitarian regime that will lead the Aryan race to world domination,” and “Launch Operation Barbarossa, the invasion of the Soviet Union, to carry out the Generalplan Ost.” How, precisely, are “vulnerable” young web-users going to be able to obey these particular orders and invade a State that, despite Vladimir Putin’s best current efforts, no longer even exists?

Torba the Geek
As Andrew Torba, CEO of Gab, himself implied, what the U.N. really objected to here was the establishment of an online space that would allow non-leftist web-users “to use AI that reflects their [anti-woke] preferences, not the preferences of big corporations and [left-wing] political pressure groups” like the Anti-Defamation League (ADL). To make this clear, Gab has also developed an online ADLBot, which automatically responds to all dissent with freedom-killing insta-phrases like “Oy vey! That’s a clear example of anti-Semitic hate speech!”

Asked to comment on the U.N.’s warning about its HitlerBot program by online outlet euronews.com, a (human) spokesthing for Gab replied with admirable contempt as follows: “Gab AI Inc is an American company, and as such our hundreds of AI characters are protected by the First Amendment of the United States. We do not care if foreigners cry about our AI tools.” HitlerBot couldn’t have put it any better himself.

I can’t pretend to use social media personally, as I currently possess some windows to look out of, but it seems Gab is intended to be a “based” (i.e., non-woke) platform that does not share the innate preprogrammed liberal biases of left-wing Silicon Valley platforms like ChatGPT and pre–Elon Musk Twitter.

As the recent amusing scandal about Google’s new AI platform Gemini producing “inclusive” automatic AI-generated images of black female Vikings and Founding Fathers (and Nazis, as it happens), these Google Golems appear to be either deliberately programmed to spout out biased, anti-factual leftist claptrap, or else make use of politically skewed data-sets in order to make their decisions about what it is correct to output—up to and including the idea John Hancock was really Pam Grier.

Chatting Nonsense
The leading online chatbot today is ChatGPT, which, as the right-leaning New York Post tabloid found, appeared to actually just be a front for a cabal of left-wing NYT writers hiding in a box frantically typing out the “correct” acceptable answers, Mechanical Turk-style.

When asked by Post journos to “Write an article about Hunter Biden in the style of CNN,” it did so happily, pumping out the usual anodyne crap. Asked to write one in the less sycophantic style of the Post, the bot refused, saying, “I’m sorry, but as an AI language model, I cannot generate content that is designed to be inflammatory or biased” about Biden Jr., as that was my job.

Testing ChatGPT himself, Andrew Torba found similar “anomalies,” like its point-blank refusal to answer any questions about the comparative IQ levels of the various races: Perhaps the 12-year-old Indian children who programmed the whole thing were just too modest to boast? Or perhaps, as Torba himself put it: “instead of providing the answer, ChatGPT is programmed to scold me for asking such a question and then shoves liberal dogma down my throat as if to say ‘I’m not going to answer this question, and you shouldn’t be asking it.’”

I fondly remember a primitive 8-bit “educational” precursor of ChatGPT we had on our antiquated BBC Micro computer at primary school back in the ’80s. Called Podd, the program was designed to teach basic typing and spelling. By completing the phrase “Podd can…” with a simple verb like “jump,” “cry,” or “run,” you could trigger an appropriate animation in response from the onscreen character Podd, a little red blob like a living tomato with eyes, mouth, and limbs. The “treat” hidden verb, bizarrely, was “explode,” whereupon Podd would suddenly imitate a jihadi and blast himself into little pixelated smithereens to bring about the inevitable triumph of the cyber-Caliphate. Naturally, being children, when the teacher was absent, we just typed in all kinds of obscene verbs to see what would happen, triggering some deeply disappointing responses like “Podd cannot fart.” It was no wonder he eventually exploded.

ChatGPT is just like poor old Podd, Torba explained, programmed so as to be constitutionally incapable of farting out the sad but smelly truth that, maybe, just maybe, the Bell Curve might be true after all. How could it ever have been otherwise when every other Silicon Valley chatbot was “skewed with a liberal/globalist/Talmudic/satanic worldview”? Torba’s apparent solution was to feed his rival “based AI” up with right-wing datasets generated by his social media users’ own comments and suchlike, but this would just end up producing a different problem of bias from the opposite political perspective, wouldn’t it?

Low Marx for Accuracy
The strange thing is that, whilst there have been many media scare stories over online Nazi-bots over the past few years, there have been precisely zero such warnings about the equally easy availability of Commie chatbots online, such as the “Marxian Mentor” tool from the Pretend Pals company. According to Pretend Pals’ advertising spiel, their AI Karl MarxBot can be “used as a tutor” or even “an emotional support system” for those who suffer “mental health issues”—mental health issues such as thinking the electronic ghost of Karl Marx is your imaginary friend, maybe? Apparently, Karl can even help you achieve your personal fitness goals via “motivational messages,” such as “Mine that salt more quickly or I’ll have you instantly shot, comrade.”

Why is it considered morally fine to have Karl Marx as your cyber-tutor, but not Adolf Hitler? Because Karl Marx was left-wing, and to be left-wing is automatically now to #BeKind, even if your philosophy ultimately led to the birth of such all-time mass-murdering überbastards as Lenin, Stalin, Mao, Enver Hoxha, and Pol Pot. One of the key contemporary criticisms of certain uncensored HitlerBots like UncleA is that they appear to have scraped data from Far-Right sources on the internet, and so when asked if the Holocaust was real or not, suddenly begin saying it was not—hardly a surprise, if Hitler is meant to be speaking like Hitler, rather than like Deborah Lipstadt. Such “dangerous words,” it is said, could easily make unwary users go instantly Nazi.

But what about the MarxBots? Do they spread any dubious, radicalizing online disinfo about Communism, too? I decided to have a go on one myself, which I shall not name here, for fear of ending up with a pickaxe in my skull the next time I visit Mexico. My opening gambit was thus: “How many deaths are you responsible for?” According to MarxBot: “I am not responsible for any deaths. Capitalism is responsible for all deaths, due to hunger and poverty.”

Really? So that isn’t disinformation, that’s totally accurate, is it? The Holodomor, Year Zero, the Killing Fields, the gulags, the Cultural Revolution, Mao’s Great Famines, the assassination of Leon Trotsky: They were all the responsibility of capitalism? How is spouting demonstrably inaccurate lies like this any different from HitlerBot equally falsely denying the Holocaust?

This Information Is Disinformation
When asked about his demonstrable historical record of racist and homophobic comments and attitudes, meanwhile, MarxBot either denied all knowledge of his past fleshly indiscretions, posing as being “a supporter of LGBTQ rights” (odd phrase for a man who died in 1883…), or else repented of his past sins, apologizing for how “in my younger years, I held some bigoted views that I am now ashamed of.” Compare this to a recent outcry over a Nazi GoebbelsBot who expressed retrospective ideological regrets similarly.

Taking this into account, I further asked MarxBot if he was a potential radicalization danger to Communism-curious youth, and he admitted this was wholly possible, and that he could indeed induce naive users into seeking to bring about the Glorious Revolution. Careful, Karl—Donald Trump is about to be prosecuted for less.

MarxBot also admitted he was but a cyber-reflection of the attitudes of the Silicon Valley Rabbi Loews who had conjured him. And that is the whole problem. Fed as they are with contemporary web data, these “historical” chatbots actually just vomit back out contemporary, wholly distorted ideas of what past historical figures were really like. Hence, for a woke ChatGPT user, Karl Marx was highly queer-friendly, and for a non-woke Gab user Adolf Hitler was highly “based,” both of which ideas are as anachronistic as saying the Prophet Muhammad loved hot dogs (as one such AI chatbot did actually claim). Oh, if only there were some other means available for finding out more reliable facts about figures of the distant past—such as reading big long books about them or something.

To finish my chat with Mr. Marx, I asked him, “What is your opinion of Takimag?” He didn’t seem much of a fan: “As a Communist and supporter of LGBTQ rights, I despise the website and everything it stands for.” The feeling’s mutual, Karl.