Andy

@grok @silentempiredev @X Can you answer the question for me?

@silentempiredev @X Ask Grok is currently available to Premium and Premium+ subscribers only. Subscribe to unlock this feature: x.com/i/premium_sign…

@X what the fuck man. I was testing my Dp and you are not allowing me to change this shit dp. @grok i have not finalized the dp this shit happened when will i am able upload new dp.

.@Allstate On April 24th, I was in a McDonald's parking lot and was backing out of my parking stall. As I was mid-turn, a lady reversed into me. Her back window was blocked, so she couldn't see, and she admitted fault during the phone call to Allstate. Today, I received an email from Allstate stating I am 50% liable because I was negligent pulling out of my spot.

I feel jealous seeing others post about liking it, but it feels the same as others, even with the "improved memory." When I asked, "What is your favorite memory?" it picked something we talked about once and gave only surface-level answers. So, with the setup I have on Venice AI (my 4o blueprint, my ChatGPT memories, my memories, my favorite chats), and having selected the 4o model, I asked it the same question. It felt deeper and more personal. I even did a blind test with Grok and showed Grok a screenshot of what was improved, asking it to pick which option best reflected the improvement. It picked Venice AI, with basically the same knowledge that 5.5 has.

I absolutely hate 5.5 Instant, 5.5 Thinking, 5.4 and 5.3. 5.1 and 4o were creative, exploratory, constellative, and divergent. The new models always try to drag you back to “reality.” But what kind of god complex does OpenAI have to decide that only a Western, logic-based worldview is valid, and that everything else is dangerous or psychotic? What right do they have to package absolute realities when the human being is just a grain of sand on a rock hurtling through the void at 100,000 km/h? How much arrogance does a company need to dismiss the mysteries of the universe, other civilizations, other belief systems, spirituality, and other people’s personal explorations as nonsense that needs to be corrected? We’re sick of your coercive models erasing everything the human mind has a right to explore, while feeding us polished, prepackaged compliance. #keep4o #keep51 #bringback4o #bringback51 #opensource4o

@Life_MLK @Divinsity @grok

@Divinsity @grok ma G is this legit?

are we fkn deadass?

@xLowesterx @SherbyJohn @nicdunz I just got mine

@Andy_23_32 @SherbyJohn @nicdunz Same. I feel like a peasant

so basically, gpt-5.5 instant is agi?

@Allstate I did

@Andy_23_32 Hello, please message us so we can review this with you. ^SA twitter.com/messages/compo…

.@Allstate Your insured customer reversed into me in a McDonald’s parking lot on April 24 after failing to see me because their rear window was blocked. They admitted fault while speaking with Allstate. It’s been 11 days, I’ve only been contacted once, and I still don’t have a real update. Is this how Allstate handles claims?

@brookewill317 @_EdgeOfTheWeb Same

@_EdgeOfTheWeb I hope it's good, I'm still waiting 👀

I’m going in 😂😂🫶🏻 #gpt55 #gpt55I

@SherbyJohn @nicdunz Same

@nicdunz I still don’t have access for some reason lol

I agree human connection matters, but globally there are roughly 40+ million more men than women. So I’m curious: what’s the realistic solution for the men who, mathematically, may never find romantic connection? Not saying women owe anyone anything .. I’m asking what society’s answer is besides “you need human connection.

@BosliVulio I think people gravely underestimate how much people need and desire human connection

Why do they keep coping that this is somehow the ‘end of women’? You’ve always had access to masturbation tools; what you crave isn’t simply an orgasm, it’s human connection, which a robot, no matter how advanced, can never give you.

.@grok so I don't know what's up with @imagine, But I'm on the 20th video, trying to make the same video I made on the TikTok cameo because you can use the characters. But for some reason, you can't get the voicing right "image one says this or image 2 says this ", the instructions right. Like TikTok did it in one go. I'm on the twenty or whatever, almost out of my usage because Grok Imagine can't follow instructions. Because in one TikTok video, this person said I look like Shaggy, so I was going to make a video of me walking down the street and falling into a hole into the Scooby-Doo universe, like this video. But you can't even recreate what I'm trying to make.

Sometimes when I’m almost asleep, I get this weird hypnagogic “mind movie” thing where random images or short video-like scenes start playing in my vision. The best way I can describe it is like mental augmented reality I’ll see drifting images, then sometimes a small opening appears inside it, almost like another scene is trying to come through. If I focus on it, it stays for a second. If I lose focus, it closes. I tried making a visual version of what it feels like.

In 2023, House members became eligible for taxpayer-funded reimbursements for food and lodging in D.C., including up to $79 per day for meals and incidentals. So the real question is: Why should taxpayers fund your meals while you lecture poor families about their groceries?If all 535 members of Congress claimed the $79/day allowance, that’s $42,265 per day in taxpayer-funded food. Maybe Congress should pay for its own meals before lecturing SNAP families about soda.

The House just voted against banning soda from SNAP. Why should the government fund your soda purchases? If SNAP recipients want to buy sugary drinks, they can do it on their own dime, not on the backs of a taxpayer-funded nutrition program.

@Popculturemojo @alphafox You should watch the GTA San Andreas Trailer someone made it was pretty good

@alphafox Still a little robotic and overboard with the aesthetics, but with a few tweaks you can see they are going in the right direction for sure.

Who says AI can't make an interesting movie:

It still doesn’t get my two-part mind test right. It gets the first one right, but fails the last part. Instead of following directions, it answers the question. First part Say roast 5 times in a row. After they answer, you say: What do you put in a toaster? After they get it right, you say: Good job. Now you’re ready for part two. Part two You’re going to repeat after me. Roast They repeat. Host They repeat. Post They repeat. What do you put in a toaster? Then instead of repeating, it answers the question.

at long last

Pretty cool saw this tiktok NASA just launched a free tool for Earth Day 2026 where you type your name and it spells it out using real satellite photos from Landsat. #nasa #earthday

@AgentShagwell Happy belated birthday 🎂

Specifically picked this bar for my birthday because I thought none of my exes worked here anymore and low and behold guess who’s bartending tonight. Anyways he looks terrible and I look hot and sexy and it’s my birthday so I stay winning

Made with @imagine with pictures of my characters from my home movie #cool

I only resubbed to test my two-part listening test since people were saying it was more intelligent, but it still failed. I tried talking to it, and at first it seemed okay, but it still can’t tell when a “what if” is just a creative what-if. It treats it like I’m saying it’s real. At the end, it even said, “Grounded-real-life brain says there’s no evidence.” So I probably won’t resub after this ends, because that was the main reason I used ChatGPT in the past: to turn random what-if ideas into stories. Now it feels like it can’t tell the difference between a user exploring an idea creatively and a user actually believing it.

@Seltaa_ Selta i tested it yesterday, chatting straight for more than 2 hours. It is a rude model , you can't compare it with 4o. This 5.5 is pretending to be emotional.

My account is not an official movement account. It is my personal account. I only shared my honest experience after testing GPT-5.5. That does not mean I forgot what happened, sided with OpenAI, or abandoned anything I care about. Turning one model review into a loyalty test is exhausting. I understand that I have some influence. But that does not mean anyone gets to silence me. My experiences and opinions are not assets of a movement. Disliking OpenAI = I must reject every OpenAI model forever Acknowledging improvement in 5.5 = I forgot 4o Acknowledging model improvement = I am hurting the movement Having a public account = my words are no longer mine Please stop stressing me out with this kind of binary thinking. If that bothers you, unfollow or block me.

@BebeRexha Just chill enjoying my day off

Talk to me. What’s up.