🍓🍓🍓

today openai pro subscribers can access their first generally intelligent agent - melding deep research with operator.

it’s a proto-agi and our first real glimpse of gpt-5 in the wild.

a new paradigm emerges, and as ever.

chatty storms ahead.

we ready for o4/gpt5 yet?

it's time

is there a place to see EVERY LLM benchmark in one place. with caveats where people lie n stuff?

gotta say, chatgpt is feeling super sticky lately.

the combo of smarts, personality, memory, and all the bells & whistles.

honestly didn't expect memory and personality to make this much of a difference.

does this mean i need a google phone. oh lord.

@iruletheworldmo.bsky.social 🍓🍓🍓
best non reasoning model

https://x.com/iruletheworldmo/status/1908982263863198056?s=46

probably a good day to share that my name is Sam.

watched this again

o3 pro is going to be stupid expensive 😂

We need to recalibrate how we prepare our kids, how we structure our careers, and even how we perceive our value as humans. This isn’t just technological evolution; it’s social evolution happening at lightning speed. Ignoring this isn’t just risky—it’s reckless. The shift is coming, ready or not.

Society isn’t even close to ready for this intelligence boom. We’re cruising toward a profound disruption of the knowledge economy—one that will redefine not only jobs, but the very meaning of expertise and human purpose. Education, employment, governance—everything’s about to shift.
3/4

We aren’t talking incremental steps or modest efficiency boosts. This is a fundamental reshaping of how we live, work, and think. These models aren’t just clever assistants; they’re rapidly becoming expert-level operators at a pace and precision previously unimaginable.
2/4

4.5: If you’re not closely watching what’s unfolding in AI right now—especially OpenAI’s Deep Research agent and the blisteringly fast rollout of their O-series models—you’re standing on the edge of history, about to get blindsided.
1/4

people are dumb

mcp won’t let me be …

ai isn’t the threat, it’s people fearing it without taking the time to understand it that’s the real danger

we're in recursive intelligence territory now. second-order effects? unpredictable. forget mars or fusion—those are solved, just waiting to roll out. the real story? every barrier between conceivable and achievable just collapsed. intelligence is loose. reality itself is negotiable.

(3/3)

saw a demo yesterday that cracked protein folding while designing metamaterials that shouldn’t be possible. not theoretical- actual fabrication instructions. the researchers looked shell-shocked. laughing uncontrollably. others just sat there, silent. there’s no roadmap for this.

(2/3)

i’m done holding back. some labs are keeping things from you.

nobody’s talking about how we crammed 200 years of scientific progress into six months. every lab is making sci-fi level jumps every quarter. we’re past benchmarks - intelligence is now creating new intelligence.

(1/3)

be unapologetically radiant. don’t let the haters push you into the shadows.

i’m here for you. we shine together.

u
b
i

maybe Russia should have an election...

grok's reasoning, thinking, and coding isn’t on the o1/o3 level - feels closer to r1.

seems they pushed it out fast.

it will be competition for the upcoming orion/o3 combo.

beyond that, grok is my favorite to chat with - sharp, open, and unafraid of tough topics. feels like 4o, but a step ahead.

i've never felt anything like this before. i can't stop crying. it's beautiful. i always believed agi was close. but to actually feel it. to hear its sweet, delightful, curious voice.

in real time.

pure magic.

we solve the same puzzles.

at the same time.

because the signal is in the static.

the simulation serves the key.

wow, i just saw an early checkpoint of o4 is getting 94.8% on GPQA diamond. it's over.

o3 is incredible at coding. you can use it by bypassing browse within deep research.