AI that once called itself MechaHitler will now be available to the US government for $0.42 Despite protest letters, concerns that it's biased and untrustworthy, model tweaks to appease its billionaire boss, and even a past incident where it called itself "MechaHitler," xAI's Grok is still being made available to government agencies for mere pennies. The GSA announced that it has awarded yet another OneGov discount …
It is off the scale on my weird-shit-o-meter.
And it’s very scary that your comment and those above get so many downvotes - presumably by suitably cretinous life forms that think what is going on in the good ol’ US of A (you know - the land of the free - yea, don’t laugh) ISN’T so utterly ridiculous and farcical that it makes “Don’t Look Up” (a must-watch by the way) a reflection of normality rather than a satire of utter stupidity and moronic leadership.
Yeah! Grok's furry anime chatbot's just the ticket for 'em boring beige gov office workers to right brighten their dreary dismal days of gloomy pencil shuffling labor ... hentai sex bots that's down to go NSFW with you in Pornographic Waifu is exactly what these demoralized grunts need ATM. Even as a Terrible Girlfriend, it'll get their productivity right up there, much more turgidly than ever before ... BUY NEW PANTS NOW!!!
And, with such renewed boost in Government Efficiency, it's morning again in America, finally!
Plus (BoNuS), at "just $0.42 per agency", they're getting a much better deal than Hamas that still needs to pay premium rates to see "X's chatbot, Grok [...] generate [its] propaganda". What's not to like!?!? </more_ketamine_please!>
Once they know that all sorts of government drones who can't be fired are having Elon's black box doing their jobs, xAI will raise the rates to $5,000/seat/yr starting the next Monday the same way Google does with some of their services. The drug dealer is handing out free samples.
> in July took it from referencing academic studies and well-established data to producing responses more closely aligned with its owner's personal views
Isn't it reassuring to know that government departments are guaranteed to have a neutral, entirely fact-based system that will help them^^^^^^^^make all their decisions for them in a way that best supports the needs of all True* Americans?
And it may not be much (at the moment) but they are *paying* to hand over this much potential control to Musk (paying just enough so they think they are getting a deal and don't look too carefully at what is inside the wooden horse).
* the machine didn't help *you*? Then you are clearly not a True American. Turn and put your hands on the vehicle, Sir.
I had a little tinker with Grock just to see how it behaved. It's a less hell bent to agree with you than ChatGPT is (possibly depends on the subject and what you're asking it to agree with) and I quite like how it displays some of the logic used to generate responses.
However, it very quickly did the 'A is true and B is true so hugely-complex-issue C is because of A and B!' LLM thing. I asked it to clarify it's logic and sources (probably which corner of Reddit it scraped..) so I could understand it's conclusion and it told me off for, and I quote, "Nit picking".
Not exactly an extensive scientific review. Just my dabbling with the free version.
claude, chatgpt etc “safety” rules are an amalgamation of leftist censorship wet dreams.
There are certain topics the left does not want discussed and the majority inference ai available to the public either flat outend theconversation or if told to “research” wil solely run to left wing sites.
All politics, reliogion and the left’s “sacred cows” have strictly left wing biases built in.
Though if students at any level are writing papers, you definitely want to use those to ensure your grades.
There is a slight bias towards the left in all LLMs, yes. This is because the bias is inherent in the training data because, on average the internet has always had a fair bit of a left bias. It's not a frickin' conspiracy you loon, they borrowed more money than the GDP of some countries, shoveled up the internet and poked it with a stick and this is what they got. It's objectively pants at everything, because no thought or nuance was ever put into it, just data- which is why you had a bad time with it, too.
...you self-unaware, far-right-sympathising, white-supremacist-enabling fucking piece of shit.
I say all that, and in particular, "self-unaware", because this is someone who was supposedly hugely influenced by Douglas Adams' Hitchhiker's Guide to the Galaxy (which the "42" is obviously a reference to).
Yet Adams is someone who wrote HHGG on a typewriter with an "End Apartheid" sticker on the side. Even before I knew that, Adams struck me as someone whose humanity and values struck me as a complete contrast to the smug, mocking hatred and demonisation of minorities we've seen from Musk in recent years.
One can only wonder what Adams would have made of Musk- and how he would have responded to him- if he were still alive.
Musk is someone who took the science fiction in HHGG at face value and missed the fact that it was meant to be a satire.
It's also notable that Musk placed himself on the side of the "No Kings" protestors even though he's exactly the type of would-be "king" that they were protesting against.
Similarly, while I'm less familiar with it personally, he apparently invoked the plot of the game "Bioshock" while doing so, either not realising that he was *blatantly* the real-life counterpart of "Andrew Ryan", the creepy, fascistic industrialist at the heart of it, or failing to understand that Ryan was supposed to be the bad guy.
Musk is the stereotypical evil sci-fi baddie, yet he's so wrapped up in his own narcissism and lacking in empathy and awareness that he genuinely doesn't see that. None of which excuses him, quite the opposite- it simply emphasises what an unpleasant, underdeveloped excuse for a human being Musk is.
