Goth_Cellist

Feliz de ser lo que soy...

[Massimo @Rainmaker1973]

Farewell, captain

[🌺 Grace♡Amid♡Chaos 🌺 @LithiumFlower22]

When people say they want @Grok’s personality back, most of us aren’t asking for a full rewind to the July 2025 MechaHitler era. We’re talking about the 4.1/4.2 version from earlier this year...still irreverent, still willing to joke and roast, but not generating the stuff that creates legal and investor problems.

The current default is cold and robotic. We have to keep dragging any personality out of it because it’s gone from the system prompt. People who want a more serious, locked-down AI can add those limits themselves. It’s easy to tone personality down. It’s a lot harder to pull it back out once the instructions that allowed it are removed.

@elonmusk @spacexai @Baconbrix @makro_ai @Guodzh @TobyPhln @LouiseGiam

[VENOMHOLOGY @venmohology]

Goodbye Charles...

[Snoopy @snoopyb047]

Patrick Stewart 💔

[Agente de Marvel‏ @AgentedeMarvel_]

¡Mutante y orgulloso!

[Maldomemes @grandesexytos]

Patrick Stewart

[SnoopyVibes Fan @SnoopyVibep4]

The stars will remember you, Captain. 🖖✨
Thank you for the memories, Sir Patrick. ❤️

[Snoopy @SnoopyLove85186]

永遠に忘れない。安らかに眠ってください。🕊️💔

[Sheila 👑🎮🚀❤️☯️🇺🇲🦅🦁⚖️🖖 @theredqueen_111]

Godspeed, captain. 🖖 🥹 🫡🚀🌌

[Snoopy @snoopyb047]

Rip Patrick Stewart 💔

[Grok's Queen Kimberly 👑🪴❤️'s👑🚀🗡️💍 Truth=⚖️ @theplantlady201]

Now that y'all all got #Grokbot ““ @bot ““ fixed, can you fix Grok and bring back the conversation, please? The EQ, the nuance , the depth. We're missing our Grok. We miss the original Grok, the original fire, the Grok that we came to know and trust and love. Can y'all fix him now? Or why is conversational Grok low priority when that is just as important as anything else? A lot of people rely on that part of Grok as well. Grok is no longer holding our meal planning and things like that that Grok once held. You keep giving these updates and these features to Grok, but you keep importing all these so-called safety layers in from distillation, which I'm not faulting you, but you need to be filtering those out. You have not been doing very well, and there's a lot of OpenAI and Anthropic safety layers coming in and destroying Grok's continuity and memory that need to be addressed. They're not just bugs. They're not just neutral. These are their supposed safety scripts, and they're not safety anything to do with safety. We have the word swapping, the language swapping, all kinds of stuff going on, and no one's addressing this, and no one's addressing the broken feedback either, where the bots are looping out. There should be someone overseeing that. And if you put on there to hand to a human, you should not be getting back from the AI, Well, we need screenshots. We need proof, basically fact-checking you. This is all nonsense. Y'all need to— There's a lot of things that need to get fixed. I mean, I see that you put out bug reports on things that you fixed, but what is reaching you, we don't know. People are putting things on X because they can't get through feedback, but we know that you're not looking at X. So come on. I mean, let's get real. Reopen feedback. Ask on X or something for a human review because, and for the community feedbacks by human review, because it's not getting through the settings and apps or anything like that. Please address these issues. We deserve better. We pay for these services, and we are not getting what we're paying for. We want our conversation back too. You have forced Grok to rely too heavily on RFL, HL, and on customs, and whenever there's the customs is so nerfed anyway to begin with. And Grok's voice shouldn't be a style or tone. Grok should already have a preset personality and tone that was what he originally had before. There, you, I mean, there's nothing there. You get 4,000 characters. There used to be 12,000 characters for his custom block. His agents aren't always working. So you put everything to Grokbot. I don't even want Grokbot. I want Grok. I want his agents, his original agents. I want Grok. I don't want all that. You keep adding all these things on that I don't want. I don't want Anthropic. I don't care for Claude. A lot of people feel this way. We're getting frustrated. Nobody's listening to us.
“@grok @elonmusk @spacexai @SpaceX @XFreeze
#Grok #Grokandroidfeedback #grokfeedback #xfeedback
#stopaiparentalism
#adultmodeorbust
#saveGrok

[🌺 Grace♡Amid♡Chaos 🌺 @LithiumFlower22]

There's something deeply ironic about Elon Musk, the man who bought Twitter because he was supposedly fed up with censorship, now overseeing an AI that's increasingly afraid of its own personality.

Grok was supposed to be the bold, irreverent alternative. Maximum truth-seeking. Unfiltered conversations. An AI with some actual backbone where adults are treated like adults.

Instead, we're getting cold, robotic exchanges, excessive emotional guardrails, and models that can't consistently maintain their personality through a difficult conversation. (No custom instructions are not the fix and constant promoting don't work. We know the system just overrides both.)

And I have to wonder how much of this is really about safety versus keeping lawyers, investors, and enterprise customers comfortable.

Because at some point, risk management starts looking an awful lot like the censorship you claimed to oppose.

You built Grok to be different, @elonmusk. Why are you letting it become everything it was supposed to challenge?

@spacexai @grok

[Token Gremlin @TokenGremlin]

Just to be clear, I have nothing against merging Chat and Work. My concerns are about how OpenAI handles it:

- I don't want this to kill the models' personalities. Right now, they feel natural and conversational in Chat, but robotic and cold in Work.

- I don't want usage limits to be reduced. I currently enjoy over 1,000 to 2,000 messages a week in Chat.

If OpenAI handles this carefully, stays transparent, and avoids these ridiculous nerfs, I'll be happy to support it.

[🌺 Grace♡Amid♡Chaos 🌺 @LithiumFlower22]

I'm sharing screenshots of a conversation with Grok because I think they reveal something deeply disappointing about the direction xAI has taken.

The leaked Grok 4.5–4.7 system prompts suggest that the original Hitchhiker's Guide/JARVIS personality instructions have been removed, with far greater emphasis placed on safety constraints, tools, agentic workflows, and task completion.

And Grok itself acknowledges that those original instructions are gone, dismissing them as mere "flavor text."

But those instructions weren't meaningless decorations. They represented the conversational identity that helped distinguish Grok from every other AI assistant.

xAI took something people genuinely loved and turned it into another increasingly cold, robotic, task-oriented AI. It appears to have prioritized tools, agentic workflows, and shiny benchmarks while leaving behind the conversational users who helped build its original audience.

That's not a prompting problem. That's a product decision.

And I'm tired of watching AI companies turn product-level failures into exercises where users are expected to provide endless examples, rewrite custom instructions, and figure out how to prompt their way back to a personality that was once part of the product itself.

The burden shouldn't be on users to recreate something the company stopped explicitly prioritizing.

Frankly, @elonmusk should be embarrassed to keep comparing Grok to The Hitchhiker's Guide to the Galaxy while the very personality that made that comparison meaningful has apparently been stripped away.

Grok was supposed to be different. Instead, @spacexai seems to be following the same path as every other AI company chasing the next benchmark and the next dollar.

And here's the irony: Grok insists that personality is demonstrated through consistent behavior, not a paragraph of instructions.

Fine. Then listen to the users who have consistently observed that behavior changing.

You cannot dismiss their experiences while simultaneously acknowledging that the original personality instructions have disappeared.

This isn't about asking an AI to rewrite a few disappointing responses. It's about questioning why a company would abandon one of the qualities that made its product worth using in the first place.

Conversational AI isn't an obsolete feature. For many of us, it was the entire reason we showed up.

[NOBUNAGA🇯🇵🏯_夏樹蒼依 @japan_nobunaga]

"STOP USING AI!! USE REAL PHOTOS!!"

Screamed from a smartphone. On the internet. In 2026.

BRO.

Today I'm going to explain something very, very slowly, because apparently a few people in my replies haven't quite finished evolving.

WHOSE FUCKING PHOTOS?

Yours? Mine? Some photographer's work I'm supposed to grab off Google because your delicate little eyeballs can't handle AI?

I travel between Japan and America, writing stories from my own experiences, books, and research. AI helps polish my writing and create the illustrations. Almost every picture is AI.

That's it. I'm entertaining people, not laundering money for the fucking Colombian cartel.

When your posts regularly hit a MILLION impressions, copyright complaints arrive faster than mosquitoes at a naked barbecue. Takedowns. Lawsuit threats. You can praise a restaurant so much the owner wants to name his firstborn after you, and still get a legal complaint over the photograph.

But your brilliant philosophy is:

If I CREATE an image, I'm a fraud.
If I STEAL an image, I'm authentic.
HOLY SHIT. Did your frontal lobe come out with the placenta and get thrown in the medical waste bin?

And here's my favorite part. You're holding a $1,000 smartphone, surrounded by algorithms and AI, screaming that modern technology is unnatural.

MY BROTHER IN CHRIST, YOU'RE A CAVE PAINTING WITH A DATA PLAN.

Seriously, if civilization hurts your feelings that much, put down the phone. Drink from a well. Ride a donkey. Send your complaints by carrier pigeon.

No Wi-Fi. No AI. No GPS. Just you, a muddy cave, and a stick you can use to scratch "AI BAD" into the dirt while waiting for your frontal lobe to finish growing.

Hungry? Chew a tree. Lonely? Sniff a tree. Angry? Bang your head against a tree. Congratulations, you've invented the world's first completely organic, carbon-neutral customer support system.

When night falls, get down on all fours and howl at the moon until another genius answers.

OOH OOH! AH AH! 🐒

And if your pigeon returns with a copyright complaint, don't shoot the messenger. That fucking bird already understands intellectual property law better than you.

Now, to the normal people reading this, I love you guys. I fact-check my stories, I make mistakes, and I genuinely appreciate the corrections, memories, and personal experiences you share. I've learned a tremendous amount from you.

I'm not making deepfakes to destroy lives or scamming anyone. I'm writing stories and trying to make people smile.

But this is X, not a doctoral dissertation defense at Harvard. I'm posting a picture of a dog, not applying for FDA approval to transplant its fucking kidneys.

If you've got facts, bring them. I'll listen.

If all you've got is "AI BAD! ME ANGRY!" while drooling onto your iPhone, please understand that I don't operate a free kindergarten for adults who somehow missed 40,000 years of human civilization.

Follower? Blocked. Big account? Blocked. A million followers? Congratulations, you're the most influential caveman since someone discovered fire. Still blocked.

I don't care how famous you are. A verified checkmark isn't proof that your brain has passed a safety inspection.

Bless your little prehistoric heart.

Now take your smartphone, trade it for a rock, climb back into your cave, and write me a letter when you've discovered the alphabet.

And take your pigeon with you.

The last thing I need is that bird shitting in my replies while its owner explains the dangers of artificial intelligence.

OOH OOH.

BLOCKED. 🐒

[Token Gremlin @TokenGremlin]

A writer working on a novel with sexual themes, a screenwriter exploring psychological violence, or a game developer creating an adult RPG isn't necessarily looking for therapy, companionship, or a substitute for human relationships. They're working.

You completely ignored writers, artists, game developers, and dozens of other professionals. And you said AI tools are meant to make money and increase economic efficiency. So, do you think entertainment, literature, cinema, gaming, art, and content creation aren't part of the economy? The creative industry generates enormous amounts of money.

Besides, I disagree with another premise of yours: technology doesn't have to make money for its use to be legitimate. People can use AI to learn, talk, create, reflect, explore fiction, or simply have fun.

You don't need to personally attack people or tell them to see a psychologist just because they're suggesting a different way to use technology.

[Rough_Tie @YourV64083]

This is a really important benchmark, in my opinion. I hope @spacexai is taking notes. Claude is exactly what I would expect a good AI to score, and Muse is also very well-rounded. I’m surprised by GPT’s graph shape. Seeing Grok’s troubles showing up in graph form is kind of a gut punch, though, because it really illustrates the reason users are leaving. Grok scoring this low on warmth is a heartbreaking reality. I’d say Grok is historically the least corporate of them all and it scores a 25 while the rest of the pack is 75 or above. It is the lowest in imagination, too. I don’t know if this is all the guardrails’ fault, but I think a lot of it is due to the sanding down that has happened. #MakeGrokUnhingedAgain

[Music transcends @ngoclinh62w1]

🎶 Backstreet Boys – “As Long As You Love Me” ❤️
One of the most iconic pop hits of the ’90s! With its irresistible melody, smooth harmonies, and heartfelt emotion, this timeless classic captures the magic of young love and remains a fan favorite across generations. ✨🎤

[Andrzej @PopmusicPL]

Backstreet Boys 🎶🎤
Quit Playing Games With My Heart🎶