That day is today! The new Grok now queries Elon Musk’s twitter feed as a solution to its antisemitic outburst problems.
I’m not kidding, I’m dead serious: https://apnews.com/article/grok-4-elon-musk-xai-colossus-14d575fb490c2b679ed3111a1c83f857
I got banned from Twitter for saying the word “cisgendered.”
So they DO know how to build limitations into the software! They’ve just done it in a fucked up, dystopian way :)
Edit: LOL I asked Grok… Cisgendered is filtered as a supposed slur, while both the R word for the mentally disabled and the hard R word for the pallor-impaired, neither of those words are filtered as slurs.
Unfortunately, you’ve been replaced by AI.
… so, the Cybertruck will blow itself up.
There’s a lot going on in there. None of it good.
As I typed it I read it and I was like “do I shoot my shot and make the joke?” and I had a rare moment of self-restraint
But Grok would be like “you had me at 𝔤𝔢𝔯𝔪𝔞𝔫”
I’m… very sad at how germane that question is in 2025.
This is Grok v4. I would demonstrate a proof, but it is the paid version, and I won’t pay Elon a bent cent.
It was released today, so this is after the shutdown where they supposedly fixed it :)
That’s the rub. Sometimes you don’t know the utility of a tool until you build it. The future is weird that way.
Who would have thought WHAT people would chat with ChatGPT about? Like, romance? WTF? That’s strange, call it emergent behavior, the final utility was unexpected.
In my head, I anticipate the tool going down the track of voice assistants from the pre-AI phase, so just like Alexa for directions or simple searches, but local to the car. Maybe unintended utilities will arise? If it can do directions and music and web searches, I guess, we don’t really know what other utilities drivers will find for it yet?
I respect your take :)
Nothing a little jumper cable jump won’t fix!
.
I was about to say, that might make the car safer on net!
Less compute means less bandwidth to hallucinate demons in front of you on a totally unobstructed straightaway.
“Oh wow they added Grok to a boomerrang, too!”
(joke only makes sense with sound on XD )
(Getting downvoted… OK :) My definition of “mildly interesting” feature may be unpopular.)
They actually do mildly interesting things, and I can see it as a vector to eventually do ACTUALLY useful things. VW and Mercedes both use ChatGPT to power this feature for now.
Plus, the Chinese ones get to wear funny hats! Here is one dressed up as 
.
“…deep analysis finding diverse sources representing ALL parties…”
Nazi party is a party. Grok is making like his forbearers by just following orders
^-- to my knowledge, this is accurate.
System prompts are the easy but wildly unpredictable way to change LLM output, but we really can’t back-trace or debug that output, we guess at what impact the s.p. edits will have.
I sort of agonized over the wording - if the system prompt is uploaded to Github, is it code, or is it documentation?
The lines are numbered like code, and I’m used to debugging software pointing out code errors by line numbers. So, code.
Don’t worry, if you’re confused, we’ll all be thrown into the same chaotic soup of coding using natural language :) With vibe coding, we’re probably already there and we just don’t feel the ramifications yet (or the endemic unemployment in IT is the ramification and we just haven’t associated the bullet wound to the loud bang yet)
The when that makes it into production:
I’m sure this is the USP that will restore sales in Canada!