The IDF has destroyed a tunnel network belonging to Hamas in the northern Gaza Strip. Early in the war, the IDF claimed to have uncovered 1,500 tunnel shafts and separate passages, destroying at least 500 alongside hundreds of distinct underground routes. Defense Minister Israel Katz stated recently that 83% of the tunnels located within the IDF-held portions of Gaza have been systematically demolished. - @MarioNawfal

"Chat with your documents" usually means uploading them to someone else's cloud. The QVAC SDK does the whole thing on-device. It ships native RAG: ragIngest your files, embed turns them into vectors locally, ragSearch finds the relevant passages, and completion answers from text the model can actually see. No external vector database. What you can build with it: a knowledge base over years of your notes, contract and research Q&A, a support bot grounded only in your manuals, a private second brain that cites its sources. Your private documents never leave the machine. No cloud vector store, no API key, no leak surface. npm install @qvac/sdk

The Egyptian Football Association released a video titled: “The Secret to Victory.” They openly declare that the key to victory is to begin each match by cursing Christians and Jews. Before every game, the team gathers to recite Quranic passages describing Christians and Jews as infidels under Allah’s wrath. Coptic Christians make up 15% of Egypt’s population, yet they are not allowed on the national team. Now, the Islamic team of Egypt has been eliminated by the ‘infidel’ nation of Argentina. Allah was not on their side.

PROGRAMMING NOTE: I just launched a chatbot on that can answer questions about my positions, built entirely from my own writing. Using the content of nearly 1,000 op-eds covering just about every political issue, and every one of them was loaded into the system, which broke the essays down into passages and built a searchable library of everything I've ever argued. Ask it anything, tariffs, election integrity, H-1B visas, and it finds the most relevant passages from my actual op-eds and builds its answer from those, citing the specific pieces so you can click through and read the full argument yourself. It doesn't guess and it doesn't put words in my mouth: if I haven't written about a topic, it tells you so. Think of it as a research assistant that has read everything I've published, remembers all of it, and is available to my readers 24/7. It is in beta and still has some bugs, but I'd love if you'd check it out and provide feedback so I can improve it and find the bugs.

SEO in 2026 isn't about ranking. It's about getting cited by AI. 68% of Google searches now end without a click. When an AI Overview shows up, organic CTR drops around 60%. But brands cited inside the AI answer get +35% CTR. What works now: 🔹 Front-load your answers. 44.2% of ChatGPT citations come from the first 30% of a document. Put a direct answer within 40 to 60 words, right after a question-form heading. 🔹 Drop CSR-only sites. GPTBot, ClaudeBot, and Perplexity don't run JavaScript. If your content is client-side rendered, AI search can't see it. Use SSR. 🔹 Write self-contained paragraphs. AI search splits pages into passages, so each paragraph has to make sense on its own. Keyword density is dead. 🔹 Name your brand in third person. "[Brand] does X" beats "we do X." AI has to identify the subject to cite you. 🔹 Brand mentions beat backlinks. Web mentions correlate 3x stronger with AI visibility than backlinks. 82% of AI citations come from earned media. 🔹 Go long-tail. 94.74% of keywords get 10 or fewer monthly searches. A wide set of clear-intent long-tail terms beats chasing high-volume heads. 🔹 Use hub-and-spoke. One pillar page plus 8 to 20 interlinked supporting articles beats a single 5,000-word guide. The fight isn't for page 1 of Google anymore. It's for a spot inside the AI answer.

A useful but little-known OpenAI API feature 📎 Users asking "where did the AI get that from?" and you can't easily show them? There's a built-in way to fix that. OpenAI's "Citation formatting" lets the model attach structured source citations to its output. A game-changer for trust and transparency in RAG and search-powered applications. 📌 Title: Citation formatting 🔗 URL: 🧩 Overview If an LLM's answer can't point to its sources, users won't trust it. Citation formatting automatically adds structured references to the model's output, showing which statements come from which source documents. This makes "evidence transparency" in RAG applications dramatically better. 🛠 How to use it Pass source documents in your input and enable citation formatting. The model generates its answer and automatically attaches references to the relevant sources. The output includes inline citation markers and structured source information, making it easy to render source links in your UI. 🏗 Building it into production ・RAG search apps: feed retrieved documents to the model, generate answers, and display source links alongside each statement. ・Internal knowledge bots: use company documents as sources and show exactly which document and section backs each answer. ・Legal and compliance tools: cite regulations and policy documents explicitly for auditability and trust. ・Education platforms: display citations from course materials so learners can check the original sources. 💡 Use cases 🔎 Cited answers in RAG applications 🏢 Evidence display for internal knowledge bases ⚖️ Legal and compliance document references 📚 Source attribution in educational content ⚠️ Watch out Citation accuracy depends on the model correctly identifying relevant passages in the sources. If sources are vague or too numerous, citation quality may drop. The quality and granularity of your source documents matter a lot. Also, citations don't guarantee 100% answer accuracy, so design your UI to let users verify the original source. ✨ "Can I trust this AI's answer?" is the biggest barrier to business adoption. Start with citation formatting in your RAG app and watch user confidence grow. #OpenAI# #LLM#

A useful but little-known OpenAI API feature 📎 Users asking "where did the AI get that from?" and you can't easily show them? There's a built-in way to fix that. OpenAI's "Citation formatting" lets the model attach structured source citations to its output. A game-changer for trust and transparency in RAG and search-powered applications. 📌 Title: Citation formatting 🔗 URL: 🧩 Overview If an LLM's answer can't point to its sources, users won't trust it. Citation formatting automatically adds structured references to the model's output, showing which statements come from which source documents. This makes "evidence transparency" in RAG applications dramatically better. 🛠 How to use it Pass source documents in your input and enable citation formatting. The model generates its answer and automatically attaches references to the relevant sources. The output includes inline citation markers and structured source information, making it easy to render source links in your UI. 🏗 Building it into production ・RAG search apps: feed retrieved documents to the model, generate answers, and display source links alongside each statement. ・Internal knowledge bots: use company documents as sources and show exactly which document and section backs each answer. ・Legal and compliance tools: cite regulations and policy documents explicitly for auditability and trust. ・Education platforms: display citations from course materials so learners can check the original sources. 💡 Use cases 🔎 Cited answers in RAG applications 🏢 Evidence display for internal knowledge bases ⚖️ Legal and compliance document references 📚 Source attribution in educational content ⚠️ Watch out Citation accuracy depends on the model correctly identifying relevant passages in the sources. If sources are vague or too numerous, citation quality may drop. The quality and granularity of your source documents matter a lot. Also, citations don't guarantee 100% answer accuracy, so design your UI to let users verify the original source. ✨ "Can I trust this AI's answer?" is the biggest barrier to business adoption. Start with citation formatting in your RAG app and watch user confidence grow. #OpenAI# #LLM#