Probably a hot take: While it’s impressive how quickly LLMs can spin up websites, I think AI is accelerating a trend that’s been happening for years: the internet is starting to look increasingly self-similar. AI tools often generate websites using the same design patterns, layouts, fonts, and aesthetics. So many new vibe-coded apps, company websites, and landing pages are starting to look identical (not to mention the content). I’m not saying AI-generated websites look bad - they often look great. But when everything looks the same, differentiation becomes more valuable. I have a feeling that as AI-generated websites become the norm, websites and online experiences with a unique aesthetic, an authentic voice, and evidence of real people behind the company will become a competitive advantage. Not just because they look different, but because they may offer better user experiences and inspire more trust. As the internet becomes flooded with AI-generated content and design, I suspect people will increasingly seek out experiences that feel genuinely human. (For as long as humans continue browsing websites themselves, at least 😅) We’re already seeing signs of this on social platforms and forums, where people are becoming increasingly skeptical of content that feels AI-generated and platforms are cracking down on AI content. I’ve also seen many recent examples of Google deindexing entirely AI-generated sites. While that doesn’t necessarily mean the AI-generated design itself is the problem, it does make me wonder how search engines will evolve their ability to distinguish between generic, mass-produced experiences and genuinely original, useful ones that happen to use AI. And I wouldn’t be surprised if search engines and LLMs increasingly favor the signals that their users are looking for - particularly originality, usefulness, and trust. (Google recently updated its documentation to emphasize human effort, originality, and skill as important considerations in evaluating content quality.) The more the internet looks like it was made by AI, the more valuable I think it will become to show evidence of human involvement. 🤔🤔🤔

Tip: select any block to see who (or what) edited it last. Now you know who to thank… teammate or bot.

We’re opening a research preview of Claude Duet, a new app where Claude works as a proactive agent. It comes to you with what needs your attention, completes tasks you ask of it, keeps working when you step away, and checks with you before it acts on your behalf.

We built ChatGPT’s ad stack to respond to an evolving conversation, while ensuring answers stay independent. Here’s how the LLM orchestrator, retrieval, ranking and auction tools work together.

OpenAI models are coming to Google Cloud. We're bringing our frontier models and Codex agentic workflows to Google Cloud's Model Garden, so enterprises can use our technology where they already build. openai.com/index/openai-mode…

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…