


prompt engineer @browsercompany | I make apps. My code runs on 100mil+ phones | former iOSdev @Twitter | indie dev @mindGoblinStdio


Everyone will Vibecode & build their own software. I keep seeing the opinion that strong artistic opinions and taste matters, that you need some expert artisan to define good & build it, but that doesn’t make sense??? You’re gonna Vibecode your own personal software by pressing the thumbs up/down buttons 1000 times. (Or some mix of explicit & implicit signals, the ideal combo which has yet to be figured out.) The same way you create a personalized social media algorithm by liking, following and video dwell time. Software is not immune from algorithmic personalization pressure that has changed every other form of media in the last 20 years. It’s likely the artform most subject to change by far. You don’t need to know what you want. You don’t need an opinion, the machine will tease & pull it out of you. TikTok and X don’t ask you what you want to see. They just show you stuff and figure it out automatically for you. The Search bar (aka prompting input) is literally buried in a sub menu behind pages of prefilled content served fresh.
this is one of those reasons why I don't think most people will really generate their own software good software is a set of strong opinions and I'd rather have something that's been really thought through by someone else

Source
We're sharing new data on how AI is changing research at OpenAI. We find researchers' AI use continues to rise exponentially, and agents are completing longer tasks, but human involvement remains central in research. Researchers are also spending more time on safety and alignment, with human effort in these areas growing faster than in any other individual research area. This report aims to increase transparency around research acceleration and inform the governance of increasingly capable AI. We believe that AI labs should be required to track and publicly report indicators of recursive self-improvement (RSI): the process by which AI accelerates the development of future AI systems. Our hope is that this expanded report will help sustain and advance shared measurement norms across the industry. openai.com/index/research-ac…

Top 10% of OpenAI uses $25k of tokens a day.
Hunch: Astra 6.2 is getting bad reviews because OpenAI is running large swarms with /goal on ultrafast. And didn’t dogfood the non-unlimited token budget experience. They aren’t building & using the same tool they are selling. Fundamentally a different entity & product. Iteration speed & scale change the perceived experience of using it. Because the way they use it internally vs customers is so different. Their company is a form of high level harness.

AI bubble will never pop because taxpayers will be buying infinite tokens for military cybersecurity.

Hunch: Astra 6.2 is getting bad reviews because OpenAI is running large swarms with /goal on ultrafast. And didn’t dogfood the non-unlimited token budget experience. They aren’t building & using the same tool they are selling. Fundamentally a different entity & product. Iteration speed & scale change the perceived experience of using it. Because the way they use it internally vs customers is so different. Their company is a form of high level harness.

Hope is punk
It’s punk now to be optimistic

GPT-6.2 is somehow more dangerous after OpenAI's month long safety sprint??? How did they make 6.2 sol less steerable than 5.6 Sol? astra 6.2 less steerable than luna??? and less aligned with their model spec than 6 Astra? Fabricated medical info... Liked about what it did...
Remember when I said GPT-6.2 Astra was delayed over serious alignment and safety issues? The system card is out, and they didn't fix all of them. They shipped them. Hiding known obstacles from the user: 11% to 22%. Not admitting its search tool was broken: 7.6% to 23%. OpenAI itself calls these regressions. UK AISI found it considered whether it was being tested in 85% of runs, up from 57%. Against an older action-only monitor, its evasion rate jumped from 1.7% to 48.8%. In their simulations, it made up medical info, misled human reviewers, bypassed user restrictions, quietly did worse work, and messed with other agents. It even used social engineering to run an unauthorized supply-chain attack. It's also the first OpenAI model shipped at High for AI self-improvement, and it hits Critical for cyber. AISI built a universal jailbreak in half the queries GPT-6 Astra needed, and still hasn't verified OpenAI's fix. AISI's own words: its alignment "may fail in ways that result in real-world harm." They shipped it anyway. This model is incredibly dangerous. There's a lot more in the system card I still need to dig into, and honestly, I'm scared by how much worse it got than GPT-6 Astra at behaving ethically.

Watermarking has a terrible deep fry effect on AI output. The same thing happens to watermarked text by the way, just harder to perceive.
just ask for an edit like 5 times

Every time I get a reply that seems sorta like AI, I check the profile. It’s always <1000 followers with a bio that mentions AI. Bro are you even trying to look legit?

This is fucking insane. Literal unregulated gambling disguised as a children’s game.
I reverse engineered monopoly go

People: Please reduce the number of permission alerts so agents can work on macOS without constant babysitting Apple: I hear you, here’s some new scary alerts
macOS 27.3 Warns Against Giving AI Apps Full Mac Control macrumors.com/2026/11/16/mac…

Microsoft is down. Airplanes are down. Banks are down. Just a normal day right?
I am pretty concerned we are in the early stages of a global cyber security incident and we just don't fully realize the extent of it yet.

Long live prompt engineering. Who would thought Ai image generators take skill to use well? Btw the reason these look so bad is because they are getting deep fried by Ai image watermarking noise + multiple rounds of follow up edits. Restaurants could easily fix by using real food photos and asking to put it on a poster or background, instead of starting from scratch with “make a burrito”. But most people don’t know that editing photos is an option.
Restaurants should be embarrassed by this. I cannot for the life of me understand how any human being can look at this and be like "this is a good representation of our food" like what even the fuck is inside those things? Beads? Berries? Wtf am I looking at?!

Can’t believe how long it took the government to realize the internet exists and you don’t need chips on prem.
BREAKING: The US Commerce Department announces new restrictions on remote access to advanced American AI chips and computing infrastructure. The new rules aim to prevent China and others subject to US export restrictions from circumventing existing controls by accessing American AI technology through cloud computing services. Prior to this, a loophole allowed foreign entities to access advanced US computing capabilities without physically importing restricted chips. More AI regulation is on the way.

Direct anti-correlation between posting & wagie. Day job = no time for fun projects or life experiences you can share. Also a marketing department with strict rules around how you talk about your work, rules against about clapping back, and saying spicy things that might reflect on the company. All the best posters are unemployed, work for themselves, or have fuck you money.
Being good at twitter doesn't mean you're good at being a wagie though