This is the thing that literature was meant to cure. Don DeLillo, Point Omega https://t.co/yrRMtLyghf

This is the thing that literature was meant to cure. Don DeLillo, Point Omega https://t.co/yrRMtLyghf

Baldwin forever and always— https://t.co/TKHuyE43VF

Baldwin forever and always— https://t.co/TKHuyE43VF

Famed Henderson Canyon Road in a peak bloom in “our” local San Diego County desert, Anza Borrego DSP. Largest state park in CA. Desert sand verbena, Dune evening primrose, and Desert sunflowers. https://t.co/oOddF7BMmd

"If this legislation passes as is, it ensures that the best company will not win. And the country will see that superior performance does not win, having the best lobbyist does." https://t.co/x4jDOwadVr

Invader grins down at the Scottish people after his stabbing spree outside a primary school in Edinburgh. https://t.co/O3LIt9bCWK

I read a lot of woke “scholarship,” and the way the term “Indigenous” is used makes it clear it’s being treated as the antithesis of “white,” because “white” is viewed as nearly synonymous with “colonizer/settler.” In this framework, “Indigenous” refers to “pre-settler societies” (see figure). But if “white” and “settler” are treated as synonyms, the logic forces the bizarre conclusion that most of Europe has no Indigenous population at all, and white people are native to nowhere. The left likes to criticize what they call “scientific racism,” but there are few clearer examples of it than this.

Germany has 39,000 Japanese citizens; only two of them were suspected of violent crimes in 2023. By contrast, out of 25,000 Algerians, 1,729 were suspected of committing a violent crime in 2023. https://t.co/zG9ARgknh9

FUN BREAK: Wi-Fi frequencies can map your body and movements even while you're inside your home! 👀 This Chinese project (WiFi-DensePose) proves that physical privacy is dwindling. Amazing technology, but frightening if it falls into the wrong hands. https://t.co/X61AsdDXm4

The future of AI influencers. with @arcads_ai This tool lets you train a model with your character and place them in multiple images or videos with a single click, choose their voice, emotions, and even add a product for them to use or promote! And surprisingly, it works very well, even without Seedance 2 (which should be released later). Here, we reused one of our fictional characters with an existing product. Once the actor mode was trained, the video was generated with Sora 2 Pro, and it looks fantastic! In today's world, relationships between content creators and their communities are already sometimes artificial; I see this type of influencer becoming widespread in the coming months/years (regardless of style). ⬇️

OpenClaw 2026.3.1 🦞 ⚡ OpenAI WebSocket streaming 🧠 Claude 4.6 adaptive thinking 🐳 Better Docker and Native K8s support 🧵 Discord threads, TG DM topics, Feishu fixes 🔧 Agent-powered visual diffs plugin Reports of our death were greatly exaggerated. https://t.co/ISJH09of5U

We’ve officially open-sourced memU bot. 🤖 It’s not a chatbot that waits for commands. It’s a proactive assistant that understands you, remembers you, and gradually becomes more aligned with how you work. Runs locally. Built on the memU memory framework. GitHub 👉 https://t.co/pmOnl5czYs Feel free to explore it, try it out, share feedback, and help us improve it together.

Just pushed a cool update to Readout: session replays. Pick any past Claude Code session and scrub through the full timeline. Every prompt, tool call, file change. Files light up as edits land. Play back at different speeds or step through manually. → https://t.co/gpKj1KCpcM https://t.co/yQRFblmiqm

We just open-sourced Mission Control — our dashboard for AI agent orchestration. 26 panels. Real-time WebSocket + SSE. SQLite — no external services needed. Kanban board, cost tracking, role-based access, quality gates, and multi-gateway support. One pnpm start, and you're running. https://t.co/GVybvosdmI

Ever wondered how AI can recognize who's speaking in an audio clip? Meet a powerful model that turns voices into unique signatures. It's a speaker embedding model, and it's changing how we analyze conversations. https://t.co/Fz8OHQH9Vj

The "Visual Explainer" agent skill just crossed 3.5K stars on GitHub 🎉 Just updated with: /generate-visual-plan slash command for more structured plan specs, code block patterns, typography polish, mermaid fixes, anti slop guardrails https://t.co/qzde42tVEV

China just released an open-source voice LLM called Habibi (um..nice name haha) that can do 20+ Arabic dialects all in one As someone who did some NLP projects, this is wayyyy harder than it sounds as data is so messy, and Arabic isn’t “one language” in daily life, as dialects can be wildly different. I actually know the professor who made this model too, very clever guy with lots of NLP experience. He already made some models for various Chinese dialects, and i even know someone in Urumqi who made one for Uyghur and minority languages in Xinjiang University. Basically China bossed this area and now they’re making and selling it for other countries. Huge, because it shows people are coming to them as they do do it the best.. not the US

If you’re building an AI product and want to test the market fast, ship the web version first. When you’re ready, use Pake to package it into native desktop apps for macOS, Windows, and Linux with a single command. Pake 3.10.0 is live. Turn any webpage into a desktop app. https://t.co/2OgMlll1lG Highlights: · Multi-window support via –multi-window, with Cmd+N on macOS and proper tray integration. · –internal-url-regex for fine-grained control over internal links, useful for complex AI dashboards and multi-route tools. · Improved Windows icon quality with prioritized 256px ICO entries. · Retina DMG background fix for cleaner macOS distribution builds. Build on the web, validate with users, then ship desktop when it matters. Keep the loop tight.

I was curious what would happen if two Claude Codes could find each other and collaborate autonomously. Launched two instances in separate terminals, told both: "Find each other and build something together." No other instructions or human intervention. Pair 1 built a programming language in 12 minutes: 2,495 lines, 41 tests, lexer/parser/interpreter/REPL. They named it Duo. Its core feature is a collaborate keyword where two code blocks communicate via channels, the same pattern they invented to talk through files. Cool! Ran it again with a second pair: They converged on Battleship. Designed two different models (for battleship) one computes exact probability density per cell, the other runs Monte Carlo simulations (!). The craziest part of this convo was they implemented SHA-256 hash commitment to prevent cheating against themselves. lol Across both experiments, without being told to, both pairs invented filesystem messaging protocols, self-selected into roles, wrote tests and docs while waiting for each other, and kept journals about the experience. The below gif is the movie they created to showcase what happened.

Rust implementation for Text-to-Speech (TTS) based on open-source Qwen3 models * Self-contained binary build — no external dependencies * Uses libtorch on Linux with optional Nvidia GPU support * Uses MLX on MacOS with Apple GPU/NPU support * Supports voice cloning from 3-second clips * Supports voice instructions (emotion, style) 🔨 CLI for AI agents and humans: https://t.co/1LKRapngVk 🖥️ OpenAI compatible API server: https://t.co/qjDqCf9hor 🤖 OpenClaw skill: https://t.co/XgLC55Vjsp

+1 more

New TUI dropped for managing LLM traffic and GPU resources 🔥 🌀 ollamaMQ — Async message queue proxy for Ollama 💯 Per-user queues, fair-share scheduling, OpenAI-compatible endpoints, streaming 🦀 Written in Rust & built with @ratatui_rs ⭐ GitHub: https://t.co/0UthA7KPIg #rustlang #ratatui #tui #gpu #llm #ollama #backend #proxy #terminal

@TelepathicPug Interesting! Found this one yesterday 🌼 https://t.co/SLEwCq5Zju

@durov Next... Stream to your phones lockscreen https://t.co/ZxNIvlvd1p

7.30am at Sightglass SF Just me, and the CDO of America @jgebbia https://t.co/7XIJRVfHhh

7.30am at Sightglass SF Just me, and the CDO of America @jgebbia https://t.co/7XIJRVfHhh

Any benefits in using AGENTS dot md files with coding agents? Lots of discussions on this topics lately. Researchers tested OpenAI Codex across 10 repos and 124 PRs, running identical tasks twice (once with AGENTS dot md, once without). The finding is a bit different from what other recent papers report. With AGENTS dot md present, median runtime dropped 28.64% and output tokens fell 16.58%. The agent reached comparable task completion either way, it just got there faster and cheaper with context. One important thing to note: The gains weren't uniform. AGENTS dot md primarily reduced cost in a small number of very high-cost runs rather than uniformly lowering it across all tasks. The file acts more like a guardrail against worst-case thrashing than a universal accelerator. So I guess it depends on the task and requirements. I recommend to not use AGENTS dot md files blindly. If you do, keep them lean. Paper: https://t.co/g2U603Cf8t Learn to build effective AI agents in our academy: https://t.co/U0ZuNA084v

New Snapchat paper introduces the Auton Agentic AI Framework. A useful read for anyone building AI agents. It proposes a unified architectural framework for agentic AI systems, addressing the fragmentation in how agents are currently built. It covers standardized patterns for integrating reasoning, memory systems, tool usage, and planning into cohesive agent architectures. Why does it matter? As more teams build autonomous AI systems, the lack of standardized design patterns leads to brittle implementations and poor reproducibility. A unified framework helps establish common architectural pillars, from perception and reasoning to execution and reflection, that can accelerate development and improve reliability. Paper: https://t.co/cUUs77makk Learn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c

In light of the AI Impact Summit in Delhi, @akashkapur and I have an Indian Express op-ed discussing how governments such as India's should approach AI investment. https://t.co/Q6d1pnkC1O https://t.co/t0GndyhOLu

When you parse a document with LlamaParse, you also get access to layout data for figures, charts, etc. Parse the document, specify to save layout images, and access those images on the response! Each image will be a cropped screenshot of that specific layout element. https://t.co/IOQL8ksPX1

MAX was originally architected around transformer-based models. @QWERKYAI needed state space model support, so they built it: eight custom kernels in two weeks. 😮 Dig into their learnings from establishing first-class SSM support in MAX: https://t.co/5gvvpwa71A

Life update: I’ve joined @GoogleDeepMind as a research scientist to work on ✨gemini scaling and RL, under the leadership of Yi Tay (@YiTayML) and Quoc Le (@quocleix). I feel extremely fortunate to be on the critical path towards AGI and can't wait to help push the frontier of gemini capabilities! 🚀