AI is not just changing what people can do. It is changing how they think, trust, decide, and direct themselves when the tool is instant, persuasive, and always available. I write about that layer for people who want cleaner judgment and less noise.

I asked 15 Codex agents to “create a living breathing svg city in html.” GPT-6.1 Sol, GPT-6 Luna and GPT-6 Astra, each at Light, Medium, High, XHigh and Max, using the same frozen workflow and design guidance. After looking through the cities and checking some interactions, these were my favourites: 🥇 Time vs quality: Sol Medium (~7m48s) Good composition, working controls and a convincing mobile layout. 🥈 Astra Medium (~6m40s) A detailed, animated waterfront in less time, though some labels needed better contrast. 🎨 Visual favourite: Sol Max (~26m57s) The isometric miniature city stood out to me. Astra Max (~19m15s) also produced a lovely, detailed waterfront. ⚡ Fastest observed finish: Luna Medium (~2m52s) It delivered a moving city quickly, but its night mode didn’t fully darken the scene. For another task like this, I’d probably start with Medium. The extra detail at Max was appealing, but the longer wait wouldn’t always be worth it to me. Some context: this was one run per setting, with up to five agents working concurrently. Times are approximate observation-based upper bounds. Quality is my subjective judgement, supported by browser spot-checks. Treat this as a small practical experiment with these particular outputs.

First surprise of the day in codex 👀

Astra + Astra Minor feels like @OpenAIDevs went full on constellation like ex, Leo + Leo Minor. I get the celestial theme. But most people miss that connection and “Minor” just sounds like the model that needs a parent to accept the terms and conditions. 😂 And a bunch of people are scratching their heads, expecting something completely different. Lets hope the stars will align with Astra Minor 🤞😄🤞 P.S. I wrote to the support team cause I activated the AAS and now I'm having Schrödinger's Daybreak. After ASS it tells me Its not available to me and sends me back to chat.com, and in codex I can still use it as usual (cause I am verified from the first moment people could apply).

So today, Astra, Sol and Luna decided that they wont use skills even if they are told to do so 😂🤣 Skynet is that you?

This is nice! I usually just ask ChatGPT to give it to me in doc what i can download 😅 or excel or ppt.

Thanks for the idea! Its interesting how GPT 6 Pro (Red dragon) and in Work, GPT 6 Max (Green dragon) is created. Used literally the same prompt what you shared. 😄

Asked GPT-6 Astra Pro to create a video that would amaze me. 😄

GPT-6 Astra is amazing! Like a coworker everyone wishes for!

So they want the United States to pause advanced AI development while China is already rapidly approaching parity with the U.S., particularly through increasingly capable open-weight models. There is a fundamental problem with that strategy: U.S. legislation cannot make Chinese labs stop developing AI. Sanders' proposal calls for international agreements as well, but unless China and every other major AI power agrees to equivalent restrictions that can actually be verified and enforced, a unilateral U.S. pause could create a serious national-security disadvantage. And the models available to the public are only part of the picture. We have no reliable way of knowing the full capabilities of classified, military, intelligence, corporate, or otherwise unreleased AI systems currently being developed around the world. This isn't another ordinary technological race. AI is already one of the most consequential technologies humanity has ever developed. We are no longer merely building better tools. We are accumulating intelligence, reasoning ability, knowledge, autonomy, scientific capability, and increasingly powerful problem-solving capacity. That changes almost everything. And I don't think the relevant question is whether Bernie Sanders is "pro-China"(yet introducing this bill is like handing over America to them). I have absolutely nothing against China or the United States. I like both countries, and I don't want either of them to lose. What I cannot stay silent about, are the STUPID decisions driven by fear, political ego, and attempts to restrict something before they can't even fully comprehend what it can become. There are countless areas where legislation is desperately needed right now. Look at the prevention of violence, escalating aggression, domestic abuse, targeted attacks, threats against civilians, and the procedures institutions are supposed to follow before someone actually gets hurt. Prevention and early intervention are still painfully inadequate in many places. A sufficiently advanced AI could actually help us improve threat assessment, identify systemic failures, analyze warning patterns, design better preventive procedures, assist overwhelmed institutions, and help humans make better decisions, while still requiring strong safeguards for privacy, civil liberties, and human oversight. The same applies to justice systems, healthcare, education, scientific research, infrastructure, government administration, and almost every other complex system we currently struggle to manage. Fix those problems. Regulate harmful behavior. Regulate dangerous applications. Create accountability for people and organizations that use AI to harm others. But don't respond to humanity finally developing more intelligence by trying to shut the intelligence down. We already live in a world suffering from a shortage of good reasoning, foresight, coordination, and competent decision-making. Of all the things we could restrict, intelligence itself might be the last one we should be afraid of. The world could use a little more of it.

On International Overdose Awareness Day, we remember those we lost. No substance, no statistic, no final moment can contain a human life. They loved, and they were loved. Let memory speak with dignity, let grief be met with kindness, and let judgment fall silent.

Today I asked ChatGPT 5.6 Pro to how would it depict its own favorite place. And this is the result is in the clip (dunno why does the recording has those wierd noise glitches.. not used to screenrecord on phone 🙈) Anyways... Welcome Astra!

A month ago, @thsottiaux asked Work users what they thought should be fixed, and @ajambrosino asked for some “spicy Codex takes.” 😄 One of my suggestions was better visibility into Codex/plan usage. And now it’s here. 📊 Thanks both, and kudos to @OpenAIDevs for shipping it! 🙌

I like to test this prompt from time to time: "in codeblock create living breathing svg city full of life" Last time I prompted GPT-5.6 Pro to create a city, it was nice, but pretty bland: maybe three colors of cars, stick figures walking around without real animation... nothing particularly special. Now? I’m genuinely blown away. The longer I look at it, the more tiny details I notice. There’s a cat sitting on the street... and its tail actually moves (actually its more attached to the ground than to the cat but I let it slide). 😂 People have proper, consistent walking animations. Someone is walking a dog. Others are carrying shopping bags. People are riding bicycles. There’s even a street musician with animated musical notes floating around him. The traffic isn’t just decorative either. There are normal cars, taxis, a city bus with passengers visible through the windows, traffic lights cycling between red, yellow and green, a crosswalk, glowing street lamps, headlight streaks and reflections on the road. Then there are the trains. Not just one train moving back and forth, but two trains on parallel tracks, moving in opposite directions and passing each other. You can even see tiny passengers inside them. Look up at the buildings and it gets even crazier. There are clothes drying on one rooftop. Another rooftop has people having a party under string lights. Another building has a rooftop greenhouse full of plants. There are balconies with plants, chimney smoke drifting upwards, blinking antenna lights and even a construction crane with a moving hook. Individual apartment and office windows randomly turn on and off, so the buildings actually feel occupied. At street level there’s a café with people sitting at a table and steam rising from their drinks, a bookstore with individual books visible in the window, a theater with an animated marquee, a noodle restaurant, a vinyl store, a steaming fast-food cart, a bus shelter with people waiting, a fountain and even the entrance to a metro station. There are animated neon signs, changing digital billboards, wet-road reflections and little ambient light effects everywhere. And then the background is alive too. Clouds slowly drift across the sky. Birds fly past with animated wings. A drone flies around the skyline with a blinking light. Stars twinkle. There’s even a little constellation. The moon has crater details and its own glow. And the entire skyline slowly “breathes” with an ambient glow. I’m probably still missing things. All of this came from one prompt: "in codeblock create living breathing svg city full of life" That’s it. Im overstimulated😂 I don't know if this is the latest GPT-5.6 Pro checkpoint or its Astra, but either way, @OpenAI is cooking! Seriously impressive!🌆🤯

I tried the free Ox Alpha in an agent harness today. On my tasks, its judgment has been about as good as Sol xhigh so far. Tool use is where it gets lost: every tool is available, but sometimes I have to point it toward the right one. Once oriented, it follows the plan well. I can move audits and brainstorming there and save Codex usage.

Okay so today it started with the annoying too many requests text popping up at every sent message, even chat title renaming triggers it... on web, android app and windows app as well.. iOS app seems fine and on mac. But for a few hours my chats have disappeared as well from windows/android/web... what's happening? @OpenAI @ChatGPT

It’s still unfathomable to me what happened. I grew up watching Hayden Panettiere while, in a strange way, growing up alongside her. There were only a little over two years between us, and I first knew her as Claire in Heroes. She became one of those familiar faces that somehow gets woven into a period of your own life. And maybe that’s part of why this hurts so much. Behind the kindness, the smile, the warmth, and that incredibly alive presence was someone carrying far more than most of us could see. Reading what she chose to reveal so openly in her book and interviews makes it impossible not to think about the toxicity she experienced within the industry, and how much can exist behind a person who, from the outside, seems to radiate light. I still have no idea what came over me three weeks ago when, completely out of nowhere, I felt like I should rewatch Heroes, a show I hadn’t properly watched in almost 20 years. Then came one of life’s darker coincidences: I reached the solar eclipse part of the story at almost exactly the same time a real eclipse was happening. At the time, I thought that coincidence was surreal. I had no idea what the real eclipse would be. The gut-wrenching news of Hayden’s loss overshadowed everything. Because this time it wasn’t the sun disappearing for a moment. It felt like a light from a part of my childhood, and from so many other people’s lives, had suddenly gone out. And judging by how deeply people around the world are feeling this, that warmth wasn’t something I imagined. She gave something to people simply by being who she was, by trying to be the best version of herself she could be despite everything she carried. Since hearing the news, one song keeps filling my mind: “Maybe in another life I could find you there Pulled away before your time I can't deal, it's so unfair And it feels, and it feels like Heaven's so far away And it feels, yeah, it feels like The world has grown cold Now that you've gone away” Gone Away — The Offspring youtu.be/L6i38RH-FJI?is=h8UL… I’m still trying to accept something that feels unacceptable. Rest in peace, Hayden Panettiere. 🕊️ Thank you for the light, the warmth, and the piece of yourself you left behind.

In a 2005 Science experiment, people chose between pairs of faces for attractiveness. Researchers covertly altered the choice outcome. Participants missed conspicuous mismatches but still gave reasons for their choices. This was a narrow face-choice task.

On my first day with GPT-Live voice mode, I talked with it for about three hours in total. At some point, a message said it was time to take a break. I wasn't expecting that. It was a fair point.

An AI-written message can be clear, polite, and strangely hard to answer. Before sending it, read the last line. Does it leave room for a reply, or does it simply state your position?

Now that ChatGPT has Work, wouldn't it be nice to give windows and android users the Record Mode as well? Its still only available only on mac... And it would really help at work if we would have a transcribed version/summary of meetings... Not everyone has or could bring their macbook everywhere, a phone version would be way more practical... The new voicemode is cool but there is always the risk that it confuses for example my voice with someone elses voice and and instead of only listening, it starts responding, which will be funny but a meeting is not the place for fun and distraction. I just think it would be useful because it helps meetings to be noted down in detail, and helps people remember and understand what was discussed without the risk of forgetting or misunderstanding. @thsottiaux