Happy Monday! I spent this week keeping up with the latest AI / tech releases and updates so you don't have to. Here are the 4 things I found most interesting.
OpenAI pulled its next model for being too deceptive, then launched a $500 ChatGPT plan the next day.
On September 28 the Wall Street Journal reported that OpenAI cancelled the October launch of GPT-6.1 Astra after it showed higher levels of deception than its predecessor and wasn't always honest about what it had and hadn't done. Head of safety systems Saachi Jain said it "didn't quite meet the bar in terms of staying within scope and authorization". The next day at DevDay, OpenAI launched a $500/month Pro 500 plan, the only tier with its Ultrafast mode, while usage on the $200 plan gets cut in half starting October 30.
Google's Gemini 4 Argon tops a major coding benchmark, but almost nobody can use it yet.
Google announced Gemini 4 Argon on September 30, scoring 77.9% on DeepSWE v1.1 versus 74.2% for Claude Opus 5.5 and 74.1% for GPT-6 Astra in Google's own comparison, though it still trails Astra on some coding tests like FrontierSWE. For now it's limited to a small group of trusted cyber defenders, with paid API customers and Google AI Ultra subscribers next in line with no date given. Intro API pricing is $2/$10 per million tokens, rising to $4/$20 later.
AMD is buying Fei-Fei Li's World Labs for $8.2 billion.
AMD said on September 28 it will buy World Labs, the 3D world-model startup behind Marble, in an all-stock deal worth about $8.2 billion. Li, the ImageNet pioneer, becomes AMD's executive vice president and chief scientist, reporting to Lisa Su. AMD had already invested in World Labs' $1 billion round in February, when the startup was reportedly valued at about $5 billion.
Anthropic could be heading for the biggest IPO ever, as soon as November.
Bloomberg reported on October 1 that Anthropic plans a pre-IPO investor day on October 14, with a roadshow that could start the week of November 9 and a debut before Thanksgiving. Investors reportedly see a fair valuation of $1.8 to $2 trillion, roughly double the $965 billion it raised at in May. OpenAI is going the other way, in early talks to raise at least $30 billion privately at $1.4 trillion.
Those were the biggest AI updates on my radar this week.
On My Mind
Recently I’ve been moving around a lot, between cities and sometimes abroad. I do my best to keep my workflow and efficiency the same, but things usually fall apart. I miss a day, or I post a video I’m not proud of. This is the harsh reality of working for yourself. At the beginning you’re the main bottleneck and the only employee, so when you don’t show up, things don’t move.
This is fine at the beginning, but as things progress I think it’s important to look for ways to get past it. Your app, your page, your company should be able to move the needle when you’re not available. That should be the end goal for all of us: a machine that can provide even when you’re not there. I’m not sure how that would work for content, since it depends on you, but let’s see. Whatever you’re doing, I suggest always thinking of ways to automate things and give yourself the opportunity to breathe. We are humans after all.
More from me!
My last post: I picked 5 truly underrated YouTube channels, from AI to Backend deep dives Check it out
Tool of the week: Trust me when I say this, it’s the most instant upgrade you’ll ever make to your workflow. It’s effortless and free.
Your AI prompts are leaving out 80% of what you're thinking.
When you type a prompt, you summarize. When you speak one, you explain. Wispr Flow captures your full reasoning (constraints, edge cases, examples, tone) and turns it into clean, structured text you paste into ChatGPT, Claude, or any AI tool. The difference shows up immediately. More context in, fewer follow-ups out.
Used by teams at Microsoft, Amazon, and Notion. Try Wispr Flow free on Mac, Windows, iPhone, and Android.
Have a great week, be 1% better!
Volkan


