Image created with OpenAI GPT-Image-1. Image prompt: vintage Sly & the Family Stone album-cover style, glowing neon marquee sign reading I’M BACK featuring black-and-white OpenAI swirl hologram; grainy retro print texture, vibrant 60s funk color palette, high-resolution
Codex is available now available to ChatGPT Plus users! We’ve also supplied our most requested feature: an option to enable internet access for the agent (with user control over HTTP methods and domains that Codex can access).”” / X https://x.com/gdb/status/1929970095427858636
Exciting news: @OpenAI’s GPT-Image-1 takes the #1 spot in the Text-to-Image Arena! 🖼️🏆 ➤ Outperforms Google’s Imagen-3.0 by 50+ points ➤ Major leap over DALL·E 3 Huge congrats to @OpenAI! 👏 https://x.com/lmarena_ai/status/1930296340648735147
OpenAI slams court order to save all ChatGPT logs, including deleted chats – Ars Technica https://arstechnica.com/tech-policy/2025/06/openai-says-court-forcing-it-to-save-all-chatgpt-logs-is-a-privacy-nightmare/
I literally built Notion Agent using MCP and OpenAI GPT-4o. It lets you take actions in your Notion workspace directly from your terminal. 100% Opensource Code with step-by-step tutorial. https://x.com/Saboo_Shubham_/status/1926827460298985628
At this point, having an obvious wrong citations in your AI generated report is a sign that you did not know to hit the Deep Research button on ChatGPT, Gemini, or Claude. The end of the fake-citation problem is good in some ways, but removes a major tell of bad AI use.”” / X https://x.com/emollick/status/1928271628447662367
ChatGPT can now read your Google Drive and Dropbox | The Verge https://www.theverge.com/news/679580/chatgpt-google-drive-dropbox-meeting-notes
ChatGPT for utilizing the context in your Google Drive:”” / X https://x.com/gdb/status/1929965894425555270
💥 Codex updates! * Codex is available to Plus users * Codex can access the internet * Codex can update PRs * Codex can be driven by voice and lots more to come.”” / X https://x.com/i/web/status/1929969441845952660
ChatGPT’s memory system is going to be the key advantage for Codex: the agent will remember where it found certain information, why it performed past steps, why it failed at some and will only improve over time. Does Cursor have anything similar to compete on this?”” / X https://x.com/i/web/status/1930005251039637836
Our company officially moved from ChatGPT Enterprise to building our own web app and chatbot for commercial electrical construction. I built this using @Replit and genuinely enjoyed how straightforward and developer-friendly it was. Now comes the fun part, connecting it to https://x.com/ai_electrician/status/1925645507533123929
.@calvinfo uses Codex when he’s on call to triage issues and prioritize bug fixes, to help him stay focused on product work. https://x.com/OpenAI/status/1923416743948570749
o3 plays Pokémon:”” / X https://x.com/gdb/status/1927955516405293496
AI Agents can now talk to any website directly. Microsoft’s NLWeb converts website data into APIs that AI agents can query as an MCP server. Works with OpenAI, DeepSeek, Gemini, Claude and other LLMs. 100% Opensource. https://x.com/Saboo_Shubham_/status/1927379307371864428
I suspect most people underestimate what o3 is capable of doing. One example: I gave it an Excel file for a small business I use for my classes & the single prompt “”identify the key assumptions here and give me a sensitivity analysis.”” It did a lot of work & gave a good answer. https://x.com/emollick/status/1927912505205989790
ChatGPT can now connect to more internal sources & pull in real-time context—keeping existing user-level permissions. Connectors available in deep research for Plus & Pro users (excl. EEA, CH, UK) and Team, Enterprise & Edu users: Outlook Teams Google Drive Gmail Linear & more”” / X https://x.com/openai/status/1930319398897889707?s=46
ChatGPT for Business Updates – YouTube https://www.youtube.com/watch?v=9lSRViLugE0
OpenAI finds more Chinese groups using ChatGPT for malicious purposes | Reuters https://www.reuters.com/world/china/openai-finds-more-chinese-groups-using-chatgpt-malicious-purposes-2025-06-05/
ChatGPT model guide for Plus users GPT-4o: daily tasks, writing mails, silly questions GPT-4.1: coding, formatting GPT-4.1 mini: no reason to use it, just use 4.1 instead GPT-4.5: writing essays, generating ideas o4-mini/-high: search, quick data analysis with uploaded files,”” / X https://x.com/i/web/status/1929617054887563492
also, today we are making a lightweight version of memory available to the free tier of chatgpt! memory has probably become my favorite feature in chatgpt; excited for us to improve this a lot over time.”” / X https://x.com/sama/status/1930007155723415560
An attempt to explain (current) ChatGPT versions. I still run into many, many people who don’t know that: – o3 is the obvious best thing for important/hard things. It is a reasoning model that is much stronger than 4o and if you are using ChatGPT professionally and not using o3 https://x.com/karpathy/status/1929597620969951434
Luca Guadagnino to Direct OpenAI-Sam Altman Movie ‘Artificial’ https://variety.com/2025/film/news/luca-guadagnono-openai-sam-altman-movie-1236417011/
Memory FAQ | OpenAI Help Center https://help.openai.com/en/articles/8590148-memory-faq
Wow they’re actually turning the OpenAI board drama into a comedy movie 🤣🤣🤣 This is either going to be really good or really bad lol https://x.com/i/web/status/1929982449720930755
codex gets access to the internet today! it is off by default and there are complex tradeoffs; people should read about the risks carefully and use when it makes sense. also, we are making in available in the chatgpt plus tier.”” / X https://x.com/sama/status/1930006856019390521
Codex is rolling out to ChatGPT Plus users today. Get started at https://x.com/i/web/status/1929957365119627520
We’re rolling out a few updates to Codex today: 1. Codex is rolling out to ChatGPT Plus users today. It includes generous usage limits for a limited time, but during periods of high demand, we might set rate limits for Plus users so that Codex remains widely available.”” / X https://x.com/OpenAIDevs/status/1929956778105811071
We’re starting to roll out a lightweight version of memory improvements to Free users. In addition to existing saved memories, ChatGPT now references your recent conversations to provide more personalized responses. https://x.com/i/web/status/1929937841905381558
ChatGPT now connects to many internal data sources:”” / X https://x.com/gdb/status/1930345346565312761
Selecting a Model Based on Stripe Conversion – A Practical Eval for Startups https://cookbook.openai.com/examples/stripe_model_eval/selecting_a_model_based_on_stripe_conversion
The Stripe eval: How @HyperwriteAI A/B tested models and chose GPT-4.1—the one that drove the most customer purchases for them: https://x.com/i/web/status/1929632332837015833
ChatGPT for meeting notes!”” / X https://x.com/gdb/status/1930345965254520864
Disrupting malicious uses of AI: June 2025 https://cdn.openai.com/threat-intelligence-reports/5f73af09-a3a3-4a55-992e-069237681620/disrupting-malicious-uses-of-ai-june-2025.pdf
If you read that “Diabolus Ex Machina” piece, worth noting the fact that AI systems cannot reliably go to links on the web and instead just make up content is a long-standing one dating back to the original ChatGPT 👇 It should be fixed because it is a common, bad failure mode.”” / X https://x.com/emollick/status/1930260717829058642
LLMs store more information than previously thought and grokking occurs once model capacity is exceeded – GPT-style models have a capacity of ~3.6 bits per parameter (prev. estimate 2 bit per parameter) – doubling precision from bfloat16 to float32 only increases model capacity https://x.com/i/web/status/1929918033541144794
Now your ChatGPT “temporary” chats may not be so temporary. A federal court ordered OpenAI to preserve all ChatGPT output logs, including deleted ones, across ChatGPT Free, Plus, Pro, and API—due to ongoing copyright lawsuits from news orgs like the NYT. 🔥 This is due to a https://x.com/rohanpaul_ai/status/1930569500140679250
Been spending my weekends designing clothes. To make it 100x easier I built a tool that turns any SSENSE link into a tech-pack template. ◆ paste any link ◆ scrapes measurements, photos ◆ Dall-E generates accurate blank sketches built w/ @lovable_dev, @OpenAI, @ScrapingBee https://x.com/cattybk/status/1926636472657903990
