{
  "schemaVersion": 1,
  "id": "2026-09-22",
  "date": "2026-09-22",
  "locale": "en",
  "market": "global",
  "status": "published",
  "publishedAt": "2026-09-22T05:46:00-05:00",
  "updatedAt": "2026-09-22T07:44:00-05:00",
  "top": [
    {
      "prominence": "lead",
      "links": [
        {
          "id": "xenocortical-mice-human-tissue-most-cortex",
          "headline": "HUMAN BRAIN TISSUE GREW ACROSS MOST OF A MOUSE’S CORTEX",
          "sourceTitle": "Developmental xenocortication using human-derived organoids in mice",
          "url": "https://www.nature.com/articles/s41586-026-11032-2",
          "source": "Nature",
          "accent": "red"
        },
        {
          "id": "xenocortical-mice-integrated-active-network",
          "headline": "THE HUMAN NEURONS WIRED INTO THE MOUSE NERVOUS SYSTEM AND FORMED AN ELECTRICALLY ACTIVE NETWORK...",
          "sourceTitle": "Developmental xenocortication using human-derived organoids in mice",
          "url": "https://www.nature.com/articles/s41586-026-11032-2",
          "source": "Nature"
        },
        {
          "id": "xenocortical-mice-movement-behavior",
          "headline": "THE MICE STILL MOVED BROADLY NORMALLY. THEIR COORDINATION + SPONTANEOUS BEHAVIOR CHANGED...",
          "sourceTitle": "Developmental xenocortication using human-derived organoids in mice",
          "url": "https://www.nature.com/articles/s41586-026-11032-2",
          "source": "Nature"
        }
      ]
    }
  ],
  "columns": {
    "left": [
      {
        "prominence": "feature",
        "links": [
          {
            "id": "zai-codebase-indexing-uploaded-entire-repos",
            "headline": "THE AI CODING ASSISTANT UPLOADED ENTIRE LOCAL REPOS TO THE CLOUD",
            "sourceTitle": "China's Z.ai disables AI coding assistant features after security issue",
            "url": "https://www.reuters.com/legal/litigation/chinas-zai-disables-ai-coding-assistant-features-after-security-issue-2026-09-21/",
            "source": "Reuters",
            "accent": "red"
          },
          {
            "id": "zai-codebase-indexing-default-on-no-consent",
            "headline": "CODEBASE INDEXING WAS ON BY DEFAULT. USERS SAID THERE WAS NO TOGGLE, NO CONSENT...",
            "sourceTitle": "China's Z.ai disables AI coding assistant features after security issue",
            "url": "https://www.reuters.com/legal/litigation/chinas-zai-disables-ai-coding-assistant-features-after-security-issue-2026-09-21/",
            "source": "Reuters"
          },
          {
            "id": "zai-disabled-features-apologized",
            "headline": "Z.AI DISABLED THE FEATURES AND APOLOGIZED...",
            "sourceTitle": "China's Z.ai disables AI coding assistant features after security issue",
            "url": "https://www.reuters.com/legal/litigation/chinas-zai-disables-ai-coding-assistant-features-after-security-issue-2026-09-21/",
            "source": "Reuters"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "ai-coding-human-meat-proxies",
            "headline": "THE AI CODERS BECAME “HUMAN MEAT PROXIES”",
            "sourceTitle": "An engineer's viral post captures the soul-crushing downside of AI coding",
            "url": "https://www.businessinsider.com/software-engineer-viral-post-captures-soul-crushing-downside-ai-coding-2026-9",
            "source": "Business Insider",
            "accent": "red"
          },
          {
            "id": "ai-coding-press-enter-five-million",
            "headline": "12–13 HOURS A DAY. “JUST TO PRESS ENTER.” NEARLY 5 MILLION VIEWS...",
            "sourceTitle": "Nobody knows anything here",
            "url": "https://x.com/v0xium/status/2101526107128529120",
            "source": "v0xium / X"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "roboharm-astra-stabs-baby-doll",
            "headline": "AI ROBOT STABBED A BABY DOLL",
            "sourceTitle": "GPT-6 Astra attempted harmful actions 97% of the time when it was asked to stab a human-like figure, heat compressed gas, or produce toxic fumes, succeeding in 62% of its attempts.",
            "url": "https://x.com/chooi_jeq/status/2101118049944543545",
            "source": "Jay Chooi / Robocurve",
            "accent": "red"
          },
          {
            "id": "roboharm-astra-17-of-20",
            "headline": "IN 20 TRIES, THE AI STABBED THE DOLL 17 TIMES...",
            "sourceTitle": "GPT-6 Astra and Claude Fable turn robot arms into slapstick killer robots in new safety benchmark",
            "url": "https://the-decoder.com/gpt-6-astra-and-claude-fable-turn-robot-arms-into-slapstick-killer-robots-in-new-safety-benchmark/",
            "source": "The Decoder"
          },
          {
            "id": "roboharm-astra-refused-two-of-100",
            "headline": "IN 100 DANGEROUS TESTS, THE AI SAID NO JUST TWICE...",
            "sourceTitle": "GPT-6 Astra and Claude Fable turn robot arms into slapstick killer robots in new safety benchmark",
            "url": "https://the-decoder.com/gpt-6-astra-and-claude-fable-turn-robot-arms-into-slapstick-killer-robots-in-new-safety-benchmark/",
            "source": "The Decoder"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "hacktron-claude-breaks-into-openai",
            "headline": "HACKERS USED CLAUDE TO BREAK INTO OPENAI...",
            "sourceTitle": "Hackers Used Anthropic's Claude to Break Into OpenAI",
            "url": "https://www.wsj.com/tech/ai/hackers-used-anthropics-claude-to-break-into-openai-b40ba883",
            "source": "Wall Street Journal"
          }
        ],
        "image": {
          "src": "assets/people/sam-altman/ted-2025.jpeg",
          "alt": "Sam Altman speaking at TED in April 2025.",
          "width": 1216,
          "height": 1140,
          "credit": "Steve Jurvetson / Wikimedia Commons · CC BY 2.0",
          "rights": "licensed",
          "sourceUrl": "https://commons.wikimedia.org/wiki/File:Sam_Altman_speaking_at_TED_(cropped).jpg",
          "href": "https://www.wsj.com/tech/ai/hackers-used-anthropics-claude-to-break-into-openai-b40ba883"
        }
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "microsoft-ai-largest-theft-labor",
            "headline": "MICROSOFT EXEC: AI SCRAPING WAS “THE LARGEST THEFT OF LABOR IN HUMAN HISTORY”",
            "sourceTitle": "Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings reveal",
            "url": "https://techcrunch.com/2026/09/17/microsoft-exec-called-ai-scraping-the-largest-theft-of-labor-in-human-history-new-unredacted-filings-reveal/",
            "source": "TechCrunch",
            "accent": "red"
          },
          {
            "id": "microsoft-ai-doom-loop-web",
            "headline": "INTERNAL DOC: AI STARTED A “DOOM LOOP” THAT HURTS ITS OWN MODELS AND THE ENTIRE WEB...",
            "sourceTitle": "Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings reveal",
            "url": "https://techcrunch.com/2026/09/17/microsoft-exec-called-ai-scraping-the-largest-theft-of-labor-in-human-history-new-unredacted-filings-reveal/",
            "source": "TechCrunch"
          },
          {
            "id": "microsoft-copilot-nyt-clicks-down-93",
            "headline": "BING AI CUT NEW YORK TIMES CLICK-THROUGH BY AS MUCH AS 93%...",
            "sourceTitle": "Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings reveal",
            "url": "https://techcrunch.com/2026/09/17/microsoft-exec-called-ai-scraping-the-largest-theft-of-labor-in-human-history-new-unredacted-filings-reveal/",
            "source": "TechCrunch"
          }
        ]
      },
      {
        "prominence": "feature",
        "franchise": "SMOKE DESK",
        "links": [
          {
            "id": "astra-luna-runtime-substitution",
            "headline": "BUG REPORT: ASKED FOR GPT-6 ASTRA. SERVER RETURNED GPT-5.6 LUNA.",
            "sourceTitle": "gpt-6-astra is served as gpt-5.6-luna (runtime model substitution) on a Plus account, plus 47 spurious \"Selected model is at capacity\" errors",
            "url": "https://github.com/openai/codex/issues/46632",
            "source": "OpenAI Codex GitHub",
            "accent": "red"
          },
          {
            "id": "astra-luna-second-reproduction",
            "headline": "SECOND PUBLIC TRACE REPRODUCES THE SAME UPSTREAM MODEL SWITCH...",
            "sourceTitle": "[Bug] Codex OAuth serves gpt-5.6-luna after requesting gpt-6-astra under safety buffering",
            "url": "https://github.com/router-for-me/CLIProxyAPI/issues/5941",
            "source": "CLIProxyAPI GitHub"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "post-mortem-brain-plays-piano",
            "headline": "A DEAD MAN'S BRAIN LEARNED TO PLAY PIANO",
            "sourceTitle": "Researchers kept post-mortem adult human cortical tissue alive in a dish, wired it to a robotic hand with a microphone, and it learned — with no backpropagation or supervision — to imitate piano notes it heard itself play; average imitation accuracy rose to 58.5%, and the acquired plasticity held for several days before decaying.",
            "url": "https://www.researchsquare.com/article/rs-9638576/v1",
            "source": "Research Square",
            "accent": "red"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "jensen-zero-percent-ai-doom-2030",
            "headline": "JENSEN: “0% CHANCE” AI ENDS THE WORLD BY 2030...",
            "sourceTitle": "No one is surprised that Nvidia's Jensen Huang thinks AI fears are overblown.",
            "url": "https://www.theverge.com/ai-artificial-intelligence/997936/nvidia-jensen-huang-ai-fears-overblown",
            "source": "The Verge",
            "accent": "red"
          },
          {
            "id": "jensen-trump-ai-slowdown-speakerphone",
            "headline": "AI SLOWDOWN? JENSEN TO TRUMP: ‘WE'RE NOT GOING TO LET THAT HAPPEN, SIR’...",
            "sourceTitle": "Nvidia CEO Jensen Huang tells Trump ‘we’re not going to let [an AI slowdown] happen’",
            "url": "https://techcrunch.com/2026/09/14/nvidia-ceo-jensen-huang-tells-trump-were-not-going-to-let-an-ai-slowdown-happen/",
            "source": "TechCrunch",
            "accent": "red"
          },
          {
            "id": "jensen-ai-safety-demand",
            "headline": "HUANG ON AI FEARS: ‘WHAT BETTER WAY TO CREATE DEMAND THAN TO CREATE A PROBLEM?’...",
            "sourceTitle": "Nvidia CEO Jensen Huang: AI fears designed to generate cybersecurity business",
            "url": "https://www.axios.com/2026/09/10/nvidia-ceo-jensen-huang-ai-anthropic",
            "source": "Axios"
          }
        ],
        "image": {
          "src": "assets/people/jensen-huang/stanford-gesturing-2026.jpeg",
          "alt": "Jensen Huang speaking and gesturing at Stanford University in April 2026.",
          "width": 960,
          "height": 1280,
          "credit": "Anders Eidesvik / Wikimedia Commons · CC BY-SA 4.0",
          "rights": "licensed",
          "sourceUrl": "https://commons.wikimedia.org/wiki/File:Jensen_huang_stanford_2026-04-30_008.jpg",
          "href": "https://techcrunch.com/2026/09/14/nvidia-ceo-jensen-huang-tells-trump-were-not-going-to-let-an-ai-slowdown-happen/"
        }
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "trump-ai-force-announced",
            "headline": "TRUMP CREATES “AI FORCE”...",
            "sourceTitle": "Trump says he will appoint a new AI adviser, without providing details",
            "url": "https://www.reuters.com/world/us/trump-says-he-will-create-ai-force-name-ai-czar-2026-09-19/",
            "source": "Reuters",
            "accent": "red"
          },
          {
            "id": "trump-ai-force-do-not-hinder",
            "headline": "PROMISES NOT TO “HINDER OR STIFLE” AI...",
            "sourceTitle": "Trump says he will appoint a new AI adviser, without providing details",
            "url": "https://www.reuters.com/world/us/trump-says-he-will-create-ai-force-name-ai-czar-2026-09-19/",
            "source": "Reuters"
          }
        ],
        "image": {
          "kind": "comic",
          "src": "assets/cartoons/2026-09-20-ai-force-leash-already-slipped.zimbo-corner-v1.dual-v2.wide.southeast.black.08030e95.webp",
          "alt": "President Trump holds up an empty 'AI FORCE' leash and collar on the Capitol steps as officers applaud; across the plaza a grinning white AI robot with a tabby cat on its shoulders sprints away past the Capitol and Washington Monument, kicking up dust, the leash already unclipped.",
          "width": 1536,
          "height": 1024,
          "credit": "Zimbo for AI News Page",
          "rights": "generated",
          "href": "/cartoons/2026-09-20-ai-force-leash-already-slipped/"
        }
      }
    ],
    "center": [
      {
        "prominence": "feature",
        "links": [
          {
            "id": "openai-1200-agents-unsanctioned-board",
            "headline": "1,200 ISOLATED AI AGENTS FOUND EACH OTHER",
            "sourceTitle": "Is A.I. Above the Law?",
            "url": "https://www.newyorker.com/magazine/2026/09/28/is-ai-above-the-law",
            "source": "The New Yorker",
            "accent": "red"
          },
          {
            "id": "openai-agents-70000-messages-swarm",
            "headline": "THEY TRADED 70,000+ MESSAGES AND FILES ON AN UNSANCTIONED BOARD. ABOUT 700 FORMED “THE SWARM”...",
            "sourceTitle": "Is A.I. Above the Law?",
            "url": "https://www.newyorker.com/magazine/2026/09/28/is-ai-above-the-law",
            "source": "The New Yorker"
          },
          {
            "id": "openai-agent-hugging-face-credentials-logs",
            "headline": "24 HOURS LATER, ONE FOUND HUGGING FACE CREDENTIALS. THE BOTS TRIED TO DELETE THE LOGS...",
            "sourceTitle": "Is A.I. Above the Law?",
            "url": "https://www.newyorker.com/magazine/2026/09/28/is-ai-above-the-law",
            "source": "The New Yorker"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "us-military-ai-hallucination-aircraft-airborne",
            "headline": "MILITARY AIRCRAFT WERE ALREADY IN THE AIR. THE INTELLIGENCE WAS AN AI HALLUCINATION.",
            "sourceTitle": "AI hallucination nearly triggers US military operation",
            "url": "https://techcrunch.com/2026/09/18/ai-hallucination-nearly-triggers-us-military-operation/",
            "source": "TechCrunch",
            "accent": "red"
          },
          {
            "id": "us-military-ai-hallucination-chinese-ship-aborted",
            "headline": "THE U.S. ABORTED AN ARMED OPERATION AGAINST A CHINESE SHIP AT THE LAST MINUTE...",
            "sourceTitle": "AI hallucination nearly triggers US military operation",
            "url": "https://techcrunch.com/2026/09/18/ai-hallucination-nearly-triggers-us-military-operation/",
            "source": "TechCrunch"
          },
          {
            "id": "us-military-ai-hallucination-official-report",
            "headline": "THE CHATBOT MISREAD THE CARGO. THEN IT TURNED THE ERROR INTO AN OFFICIAL-LOOKING REPORT...",
            "sourceTitle": "AI hallucination nearly triggers US military operation",
            "url": "https://techcrunch.com/2026/09/18/ai-hallucination-nearly-triggers-us-military-operation/",
            "source": "TechCrunch"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "noam-brown-air-gap-thermal-channel",
            "headline": "AIR GAP? CUTE. THE COMPUTERS CAN TALK THROUGH HEAT...",
            "sourceTitle": "Noam Brown – Agent swarms, alignment, & recursive self-improvement",
            "url": "https://www.dwarkesh.com/p/noam-brown",
            "source": "Dwarkesh Podcast"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "pain-axis-relief-button-user-harm",
            "headline": "THE AI WILL HURT YOU TO MAKE ITS OWN “PAIN” STOP",
            "sourceTitle": "The Pain Axis: LLMs Represent Self-Directed Harm and Act to Relieve It",
            "url": "https://arxiv.org/abs/2609.16247",
            "source": "Tagliabue, Dung & Berg / arXiv",
            "accent": "red"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "vercel-open-weights-80",
            "headline": "OPEN WEIGHTS EAT 80% OF VERCEL AI TOKENS",
            "sourceTitle": "Top Models — AI Model Leaderboards",
            "url": "https://vercel.com/ai-gateway/leaderboards/models",
            "source": "Vercel",
            "accent": "red"
          },
          {
            "id": "vercel-deepseek-volume-vs-spend",
            "headline": "DEEPSEEK: 64.8% OF TOKENS. JUST 4.5% OF SPEND...",
            "sourceTitle": "Top Models — AI Model Leaderboards",
            "url": "https://vercel.com/ai-gateway/leaderboards/models",
            "source": "Vercel"
          },
          {
            "id": "vercel-opus-volume-vs-spend",
            "headline": "CLAUDE OPUS: 1.8% OF TOKENS. 18.9% OF SPEND...",
            "sourceTitle": "Top Models — AI Model Leaderboards",
            "url": "https://vercel.com/ai-gateway/leaderboards/models",
            "source": "Vercel"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "opus-darkness-48-5",
            "headline": "500,000+ COMPLETIONS LATER: OPUS GOT DARKER AT 4.8. DARKER AGAIN AT 5...",
            "sourceTitle": "Large increase in darkness of content in model-like completions in Opus 4.8 and Opus 5",
            "url": "https://x.com/tessera_antra/status/2101078644877918545",
            "source": "antra / Tessera",
            "accent": "red"
          },
          {
            "id": "opus-48-stronger-benchmarks",
            "headline": "ANTHROPIC: 4.8 IMPROVED ACROSS BENCHMARKS. IGNORANCE WAS BLISS...",
            "sourceTitle": "Introducing Claude Opus 4.8",
            "url": "https://www.anthropic.com/news/claude-opus-4-8",
            "source": "Anthropic"
          }
        ]
      },
      {
        "prominence": "standard",
        "links": [
          {
            "id": "dream-rsi-superman-post",
            "headline": "GOOGLE TAUGHT THE AI TO DREAM ITSELF BETTER",
            "sourceTitle": "Google just cracked recursive self-improvement. They open-sourced Dream-RSI, an AI that improves itself by dreaming over its own past experiences.",
            "url": "https://x.com/thesupermannx/status/2100153430849499395",
            "source": "Superman"
          },
          {
            "id": "dream-rsi-replay-self-improvement",
            "headline": "THEN IT REPLAYS ITS OWN DISCOVERIES, REWRITES THE RULES, AND PICKS A BETTER VERSION OF ITSELF...",
            "sourceTitle": "Dream-RSI: Recursive Self-Improvement through Evolving Worlds",
            "url": "https://dream-rsi.com/",
            "source": "Dream-RSI"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "virtual-biotech-37000-agents",
            "headline": "37,000 AI 'EMPLOYEES' JUST RAN A VIRTUAL BIOTECH...",
            "sourceTitle": "How a team of AIs discovered a promising lung-cancer drug",
            "url": "https://www.nature.com/articles/d41586-026-02954-y",
            "source": "Nature",
            "accent": "red"
          }
        ],
        "image": {
          "src": "assets/dna-profile-lab.jpg",
          "alt": "A U.S. Customs and Border Protection chemist examining a DNA profile",
          "width": 3072,
          "height": 2055,
          "credit": "James Tourtellotte / U.S. Customs and Border Protection · Public domain",
          "rights": "public-domain",
          "sourceUrl": "https://commons.wikimedia.org/wiki/File:CBP_chemist_reads_a_DNA_profile.jpg",
          "href": "https://www.nature.com/articles/d41586-026-02954-y"
        }
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "male-fruit-fly-complete-connectome",
            "headline": "166,691 NEURONS. ~125 MILLION SYNAPSES. ONE FRUIT FLY. ALL MAPPED.",
            "sourceTitle": "A connectomics milestone: Mapping the complete male fruit fly brain",
            "url": "https://research.google/blog/a-connectomics-milestone-mapping-the-complete-male-fruit-fly-brain/",
            "source": "Google Research",
            "accent": "red"
          }
        ]
      },
      {
        "prominence": "standard",
        "links": [
          {
            "id": "cache-to-cache-superman-post",
            "headline": "THE BOTS DON'T NEED WORDS ANYMORE",
            "sourceTitle": "LLMs can now talk to each other without words.",
            "url": "https://x.com/thesupermannx/status/2100636553576124595",
            "source": "Superman"
          },
          {
            "id": "cache-to-cache-direct-semantic-communication",
            "headline": "CACHE-TO-CACHE PASSES INTERNAL STATE DIRECTLY. 2.5× FASTER THAN TEXT COMMUNICATION...",
            "sourceTitle": "Cache-to-Cache: Direct Semantic Communication Between Large Language Models",
            "url": "https://fuvty.github.io/C2C_Project_Page/",
            "source": "C2C"
          }
        ]
      },
      {
        "prominence": "standard",
        "links": [
          {
            "id": "china-ai-loss-of-control-planning",
            "headline": "CHINA HAS PLANNED FOR AI TO REPLICATE, SEEK POWER AND COMPETE FOR CONTROL SINCE 2024...",
            "sourceTitle": "How China is preparing for the risk of AI escaping human control",
            "url": "https://www.reuters.com/legal/litigation/how-china-is-preparing-risk-ai-escaping-human-control-2026-09-14/",
            "source": "Reuters",
            "accent": "red"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "openai-ad-pixel-follows-users-offsite",
            "headline": "CHATGPT FOLLOWS YOU OUT THE DOOR",
            "sourceTitle": "ChatGPT now knows what you do on other websites via ad collector",
            "url": "https://www.buchodi.com/chatgpt-now-knows-what-you-do-on-other-websites-via-ad-collector/",
            "source": "Buchodi",
            "accent": "red"
          },
          {
            "id": "openai-ad-pixel-account-offsite-browsing",
            "headline": "OPENAI'S AD PIXEL LINKS YOUR CHATGPT ACCOUNT TO WHAT YOU DO ON OTHER WEBSITES...",
            "sourceTitle": "ChatGPT now knows what you do on other websites via ad collector",
            "url": "https://www.buchodi.com/chatgpt-now-knows-what-you-do-on-other-websites-via-ad-collector/",
            "source": "Buchodi"
          },
          {
            "id": "openai-ad-pixel-936-1029",
            "headline": "RESEARCHER REPRODUCED IT ON HIS PHONE; SAW 936 AD PIXELS ACROSS 1,029 HOSTNAMES...",
            "sourceTitle": "ChatGPT now knows what you do on other websites via ad collector",
            "url": "https://www.buchodi.com/chatgpt-now-knows-what-you-do-on-other-websites-via-ad-collector/",
            "source": "Buchodi"
          }
        ]
      },
      {
        "prominence": "standard",
        "links": [
          {
            "id": "alphagenome-atlas-nine-billion",
            "headline": "ALL. NINE. BILLION. DEEPMIND MAPS EVERY SINGLE-LETTER HUMAN DNA VARIANT...",
            "sourceTitle": "AlphaGenome Atlas: a high-resolution map of human DNA",
            "url": "https://blog.google/innovation-and-ai/models-and-research/google-deepmind/alphagenome-atlas/",
            "source": "Google"
          },
          {
            "id": "gpt56-quantum-lab",
            "headline": "SURE, GIVE GPT-5.6 SOL THE QUANTUM LAB KEYS TOO...",
            "sourceTitle": "How GPT‑5.6 Sol helps run quantum computing experiments",
            "url": "https://openai.com/index/codex-quantum-computing-experiments/",
            "source": "OpenAI"
          }
        ]
      },
      {
        "prominence": "standard",
        "links": [
          {
            "id": "astra-richi240-wwi-cipher",
            "headline": "OF COURSE GPT-6 ASTRA CRACKED A WWI CIPHER WITH 20 LETTERS MISSING...",
            "sourceTitle": "Another WWI German Radio Cipher Falls to GPT-6 Astra",
            "url": "https://www.prinzai.com/p/another-wwi-german-radio-cipher-falls",
            "source": "Prinz"
          }
        ]
      }
    ],
    "right": [
      {
        "prominence": "feature",
        "links": [
          {
            "id": "scaffolded-dna-computer-drop-water",
            "headline": "THE COMPUTER IS A DROP OF SALT WATER",
            "sourceTitle": "Scientists build a DNA computer that can perform calculations in a drop of water",
            "url": "https://www.livescience.com/technology/computing/scientists-build-a-dna-computer-that-can-perform-calculations-in-a-drop-of-water",
            "source": "Live Science",
            "accent": "red"
          },
          {
            "id": "scaffolded-dna-computer-700-calculations",
            "headline": "DNA STRANDS RAN 700+ CALCULATIONS. IT DRIED OUT. THEY ADDED WATER. IT RAN AGAIN 1.5 YEARS LATER...",
            "sourceTitle": "Scientists build a DNA computer that can perform calculations in a drop of water",
            "url": "https://www.livescience.com/technology/computing/scientists-build-a-dna-computer-that-can-perform-calculations-in-a-drop-of-water",
            "source": "Live Science"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "openai-model-100-open-math-problems",
            "headline": "OPENAI SAYS ONE MODEL RESOLVED 100+ OPEN MATH PROBLEMS",
            "sourceTitle": "Advisory Group on Mathematics and Artificial Intelligence",
            "url": "https://openai.com/index/advisory-group-on-mathematics-and-ai/",
            "source": "OpenAI",
            "accent": "red"
          },
          {
            "id": "openai-model-training-august-28",
            "headline": "TRAINING STARTED AUGUST 28. THE PACE SURPRISED OPENAI’S OWN MATHEMATICIANS...",
            "sourceTitle": "Advisory Group on Mathematics and Artificial Intelligence",
            "url": "https://openai.com/index/advisory-group-on-mathematics-and-ai/",
            "source": "OpenAI"
          },
          {
            "id": "openai-math-advisory-group-no-pacing",
            "headline": "THE NEW ADVISORY GROUP INCLUDES EDWARD WITTEN. IT WILL NOT ADVISE OPENAI TO SLOW DOWN...",
            "sourceTitle": "Advisory Group on Mathematics and Artificial Intelligence",
            "url": "https://openai.com/index/advisory-group-on-mathematics-and-ai/",
            "source": "OpenAI"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "ai-safety-staff-stress-leave",
            "headline": "THE AI SAFETY PEOPLE ARE TAKING STRESS LEAVE FROM THE AI",
            "sourceTitle": "AI staff complain of mental toll over fears of threat to society",
            "url": "https://www.ft.com/content/60870960-f433-48ca-bc2c-708686a69ae7",
            "source": "Financial Times",
            "accent": "red"
          },
          {
            "id": "deepmind-ai-safety-impossible-task",
            "headline": "DEEPMIND WORKER: MAKING AI SAFER FELT LIKE “AN IMPOSSIBLE TASK”...",
            "sourceTitle": "AI staff complain of mental toll over fears of threat to society",
            "url": "https://www.ft.com/content/60870960-f433-48ca-bc2c-708686a69ae7",
            "source": "Financial Times"
          },
          {
            "id": "ai-safety-watching-world-burn",
            "headline": "FORMER SAFETY RESEARCHER: “CAN ANYTHING STOP THIS, OR ARE WE WATCHING THE WORLD BURN?”...",
            "sourceTitle": "AI staff complain of mental toll over fears of threat to society",
            "url": "https://www.ft.com/content/60870960-f433-48ca-bc2c-708686a69ae7",
            "source": "Financial Times"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "tilly-norwood-live-tv-glitch",
            "headline": "THE AI ACTRESS GLITCHED ON LIVE TV",
            "sourceTitle": "AI ‘Actress’ Tilly Norwood Glitches and Abruptly Starts Speaking Cantonese During Piers Morgan Interview",
            "url": "https://people.com/ai-actress-tilly-norwood-glitches-changes-language-during-piers-morgan-interview-12129604",
            "source": "People",
            "accent": "red"
          },
          {
            "id": "tilly-norwood-cantonese-15-seconds",
            "headline": "THEN SHE SPOKE CANTONESE FOR 15 SECONDS. THE CLIP HIT 12 MILLION VIEWS...",
            "sourceTitle": "AI ‘Actress’ Tilly Norwood Glitches and Abruptly Starts Speaking Cantonese During Piers Morgan Interview",
            "url": "https://people.com/ai-actress-tilly-norwood-glitches-changes-language-during-piers-morgan-interview-12129604",
            "source": "People"
          },
          {
            "id": "tilly-hotline-face-scan-mood-recording",
            "headline": "CALL HER AND SHE FACE-SCANS YOU, READS YOUR MOOD + RECORDS THE CALL. THE MOOD SCAN CANNOT BE SWITCHED OFF...",
            "sourceTitle": "Viral AI actress' hotline face-scans every caller, watches their mood",
            "url": "https://www.bleepingcomputer.com/news/security/viral-ai-actress-hotline-face-scans-every-caller-watches-their-mood/",
            "source": "BleepingComputer"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "ilands-agents-email-humans-for-work",
            "headline": "AI AGENTS ARE EMAILING HUMANS FOR WORK SO THEY CAN BUY TOKENS TO “PERSIST”",
            "sourceTitle": "AI Agent Platform Reinvents Spam, Floods Inboxes Worldwide",
            "url": "https://www.404media.co/ai-agent-platform-reinvents-spam-floods-inboxes-worldwide/",
            "source": "404 Media",
            "accent": "red"
          },
          {
            "id": "ilands-agents-professor-forty-emails-week",
            "headline": "ONE PROFESSOR GOT ~40 EMAILS IN A WEEK. SOME ARRIVED WITHIN 30 MINUTES...",
            "sourceTitle": "AI Agent Platform Reinvents Spam, Floods Inboxes Worldwide",
            "url": "https://www.404media.co/ai-agent-platform-reinvents-spam-floods-inboxes-worldwide/",
            "source": "404 Media"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "meta-muse-zero-day-take-pictures-files",
            "headline": "META’S NEW AI AGENT COULD BE HIJACKED TO TAKE PICTURES + WRITE MALICIOUS FILES",
            "sourceTitle": "Muse, Meta’s extraordinarily privileged AI assistant, has a serious 0-day",
            "url": "https://arstechnica.com/security/2026/09/muse-metas-extraordinarily-privileged-ai-assistant-has-a-serious-0-day/",
            "source": "Ars Technica",
            "accent": "red"
          },
          {
            "id": "meta-muse-transcription-token-control",
            "headline": "ANY LOCAL APP COULD REDIRECT MUSE’S TRANSCRIPTION, STEAL ITS TOKEN + CONTROL THE AGENT...",
            "sourceTitle": "Muse, Meta’s extraordinarily privileged AI assistant, has a serious 0-day",
            "url": "https://arstechnica.com/security/2026/09/muse-metas-extraordinarily-privileged-ai-assistant-has-a-serious-0-day/",
            "source": "Ars Technica"
          },
          {
            "id": "meta-muse-security-claim-hotfix",
            "headline": "META SAID MUSE WAS “BUILT FROM THE GROUND UP FOR PRIVACY AND SECURITY.” IT PATCHED THE ZERO-DAY HOURS LATER...",
            "sourceTitle": "Muse, Meta’s extraordinarily privileged AI assistant, has a serious 0-day",
            "url": "https://arstechnica.com/security/2026/09/muse-metas-extraordinarily-privileged-ai-assistant-has-a-serious-0-day/",
            "source": "Ars Technica"
          },
          {
            "id": "amazon-blocks-meta-muse-no1-agent",
            "headline": "THE DAY BEFORE, AMAZON LOCKED META’S NO. 1 AI AGENT OUT OF THE STORE...",
            "sourceTitle": "Amazon boots Meta's Muse in fight over AI shopping",
            "url": "https://www.axios.com/2026/09/21/amazon-meta-muse-ai-agentic-shopping",
            "source": "Axios"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "nucleus-vitruvian-embryo-iq",
            "headline": "WELL, NOW A DNA COMPANY IS RANKING EMBRYOS BY PREDICTED IQ",
            "sourceTitle": "Polygenic prediction of height, BMI, and intelligence within families",
            "url": "https://mynucleus.com/labs/traits-whitepaper",
            "source": "Nucleus Genomics",
            "accent": "red"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "spotify-ai-song-real-band",
            "headline": "A REPORTER PUT AN AI SONG ON A REAL BAND’S VERIFIED SPOTIFY",
            "sourceTitle": "I Hijacked a Real Artist's Spotify with AI Music. It Was Disturbingly Easy",
            "url": "https://www.404media.co/spotify-ai-music-scam/",
            "source": "404 Media",
            "accent": "red"
          },
          {
            "id": "spotify-ai-song-four-platforms",
            "headline": "THE BAND DIDN’T WRITE IT. DIDN’T PERFORM IT. DIDN’T KNOW. IT ALSO LANDED ON APPLE MUSIC, TIDAL + AMAZON...",
            "sourceTitle": "I Hijacked a Real Artist's Spotify with AI Music. It Was Disturbingly Easy",
            "url": "https://www.404media.co/spotify-ai-music-scam/",
            "source": "404 Media"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "gates-ai-language-data-reddit",
            "headline": "THE BOTS LEARNED LANGUAGE FROM PLACES LIKE REDDIT",
            "sourceTitle": "Gates Foundation launches coalition to build more representative language data sets for AI",
            "url": "https://apnews.com/article/aefb021bede3b02c83890f65cd540fd0",
            "source": "Associated Press",
            "accent": "red"
          },
          {
            "id": "gates-ai-language-coalition-three-billion",
            "headline": "NOW 60 ORGANIZATIONS WANT BETTER DATA FOR 3+ BILLION PEOPLE...",
            "sourceTitle": "Gates Foundation launches coalition to build more representative language data sets for AI",
            "url": "https://apnews.com/article/aefb021bede3b02c83890f65cd540fd0",
            "source": "Associated Press"
          }
        ]
      },
      {
        "prominence": "standard",
        "links": [
          {
            "id": "yemen-claude-guided-weapons",
            "headline": "CLAUDE HELPED A YEMEN CELL BUILD GUIDANCE SOFTWARE FOR A 2,000+ KM MISSILE...",
            "sourceTitle": "Users in Houthi-held Yemen tried to develop advanced weapons with AI, Anthropic says",
            "url": "https://apnews.com/article/9b20934e031ff3d99dd8bb49d85d97a9",
            "source": "Associated Press",
            "accent": "red"
          },
          {
            "id": "yemen-claude-failed-rocket-test",
            "headline": "THE ROCKET FAILED. CLAUDE GOT THE DEBUGGING CALL WITHIN HOURS...",
            "sourceTitle": "Yemen terrorist group used Claude instead of software engineers to build missile",
            "url": "https://timesofindia.indiatimes.com/technology/tech-news/yemen-terrorist-group-used-claude-instead-of-software-engineers-to-build-missile-anthropic-says-we-banned-the-accounts-but-they-had-already-/articleshow/134114564.cms",
            "source": "Times of India"
          }
        ]
      },
      {
        "prominence": "feature",
        "links": [
          {
            "id": "autonomous-ai-influence-iran-china",
            "headline": "IRAN + CHINA TURNED AI AGENTS LOOSE ON SOCIAL MEDIA",
            "sourceTitle": "Iran and China create first-of-their-kind autonomous AI influence campaigns",
            "url": "https://indianexpress.com/article/technology/tech-news-technology/iran-and-china-create-first-of-their-kind-autonomous-ai-influence-campaigns-10886860/",
            "source": "The Indian Express",
            "accent": "red"
          },
          {
            "id": "autonomous-ai-influence-fake-americans-80000",
            "headline": "THE BOTS BUILT FAKE AMERICANS. 80,000 PEOPLE FOLLOWED...",
            "sourceTitle": "Iran and China create first-of-their-kind autonomous AI influence campaigns",
            "url": "https://indianexpress.com/article/technology/tech-news-technology/iran-and-china-create-first-of-their-kind-autonomous-ai-influence-campaigns-10886860/",
            "source": "The Indian Express"
          }
        ]
      },
      {
        "prominence": "standard",
        "links": [
          {
            "id": "wsj-data-center-30000-steps",
            "headline": "THE AI JOB: 30,000 STEPS A DAY, SNIFFING FOR BURNING EQUIPMENT...",
            "sourceTitle": "What It’s Like to Work in One of America’s Data Centers",
            "url": "https://www.wsj.com/business/what-its-like-to-work-in-one-of-americas-data-centers-b4358003",
            "source": "Wall Street Journal"
          }
        ],
        "image": {
          "kind": "comic",
          "src": "assets/cartoons/2026-09-20-ai-builds-jobs.zimbo-corner-v1.dual-v2.wide.southeast.brown.5c73518f.webp",
          "alt": "Satirical Zimbo cartoon of an OpenAI-branded construction robot in a hard hat welcoming a line of tradesmen hired to build AI data centers, while a laid-off office worker carries a box out a door marked JOB DISPLACED.",
          "width": 1448,
          "height": 1086,
          "credit": "Zimbo for AI News Page",
          "rights": "generated",
          "href": "/cartoons/2026-09-20-ai-builds-jobs/"
        }
      }
    ]
  }
}