We live in an era where a single prompt can trigger another Y2K. 🧯
#CyberSecurity #PHP #ai #AgenticAI
OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites.
Over the past two months, we led a coalition of maintainers, researchers, and security providers to prepare fixes and protections ahead of disclosure. This duration was necessary because the fix was quite intricate and required a major change to the PHP engine. Every site using PHP can now safely upgrade their version to stay secure, in addition to applying mitigations from various security vendors.
Due to the severity of these flaws and PHP’s widespread use, we consider this bug to be one of the most impactful vulnerabilities in recent memory. It also marks a new phase, where AI can find security bugs with internet-scale impact.
This is an example of the Defender’s Window in action: to our knowledge, frontier AI was able to find this vulnerability ahead of attackers, giving defenders time to prepare.
openai.com/index/coordinated…
Readers added context they thought people might want to know
OpenAI may claim a role in the coordinated response and patching process. But the initial discovery and exploit development, however, occurred several months earlier by an independent researcher.
x.com/lcfr_hax/statu…
Aniket Rana retweeted
We’ve committed to having evaluators assess our safety practices. We're in discussions with multiple organizations, and have now given employee-like access to a team from @METR_Evals.
In this post, we describe how we work with third-party evaluators: anthropic.com/news/an-update…
Aniket Rana retweeted
how it feels to interact with my human family after another 14 hour day of interacting with Claude
Aniket Rana retweeted
New on our Engineering blog: Creating in-house source control.
In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale.
Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: anthropic.com/engineering/re…
Spent 6 years optimizing for the H-1B → Green Card pipeline.
Turns out it had a single point of failure.
indianexpress.com/article/ex…
#PERM #H1B #GreenCard
Aniket Rana retweeted
Thrilled to welcome @spacexai as a Founding Corporate Patron for the Omacom Foundation. That's $1,500,000 in @grok tokens for the maintenance and development of Omarchy. Maybe soon they'll be beamed straight from orbit? Let's go 🚀🌕 omarchy.org/news/2026/10/spa…
Aniket Rana retweeted
I really like this image!
It perfectly reflects how I’m distributing each task across the different models
Aniket Rana retweeted
Nana Patekar is no more.
I have lost not merely an exceptional artist, but an extraordinary human being and a dear friend , someone whose warmth and affection touched everyone around him.
We spoke just eight days ago. I could never have imagined that it would be our last conversation.
His last words to me were: “Love you, Bala.”
Those words will stay with me for the rest of my life.
Some people leave behind memories. A few leave behind a part of themselves in our lives.
Nana was one of them.
Goodbye, my friend. You will be deeply missed, and forever remembered.
Aniket Rana retweeted
The execution speed will come from how fast the decisions get made, not from working longer hours.
Aniket Rana retweeted
i went to aadhaar website today, and i found a miracle 🤯 just look at the UI/UX..
after a decade feels like indian government portals have been taken over by a private company
AI has finally arrived to fix this, thanks @OpenAI @AnthropicAI
hopefully all the other indian gov websites are polishing along
Aniket Rana retweeted
@McKinsey @BainandCompany
Are there any near future plans to launch an MCP using which we can access and search across the proprietary research papers or data insights published by you guys?
I would subscribe for it and let my AI Agents read through your research directly.
I grepped my own agent transcripts to see how bad it was.
113 files out of 581 had a credential sitting in them. Plaintext, in a directory I'd never opened.
Wrote up what I found, and the hook I built afterwards.
aniumbott.medium.com/113-of-…
#ClaudeCode #DevTools #Cybersecurity
CXMT lists July 27: China's biggest-ever chip IPO, ~$85B.
A DRAM maker that didn't exist in 2016, breaking a 3-company global memory cartel. Every device you ship runs on this layer. Worth watching.
scmp.com/tech/article/336061…
#ChipWar #Semiconductors
