https://themodelwire.com/article/simile-ai-scales-generative-agents-into-population-level-behavior-models-01M0JYSE9PQHJFE2YBEDS7EVAG Modelwire en 2026-08-21T19:50:24.000Z Simile AI scales generative agents into population-level behavior models https://themodelwire.com/article/nvidia-research-elevates-agent-harness-over-base-model-quality-01M0JXXX5DQ59EXVHH14985D6G Modelwire en 2026-08-21T19:43:39.000Z Nvidia research elevates agent harness over base model quality https://themodelwire.com/article/anthropic-deploys-mythos-5-into-production-security-scanning-and-critical-infras-01M0JXXZA63S164CDCAMW4RNR3 Modelwire en 2026-08-21T19:35:37.000Z Anthropic deploys Mythos 5 into production security scanning and critical infrastructure defense https://themodelwire.com/article/deepseeks-vision-model-matches-opus-4-8-on-agent-benchmarks-01M0JW715153VZJWBAFHYVNJY1 Modelwire en 2026-08-21T19:08:27.000Z Deepseek's vision model matches Opus 4.8 on agent benchmarks https://themodelwire.com/article/american-opposition-to-data-centers-jumps-to-75-percent-in-one-year-01M0JSMMCG0W234E6EQXZZBXVR Modelwire en 2026-08-21T18:25:44.000Z American opposition to data centers jumps to 75 percent in one year https://themodelwire.com/article/coding-agents-make-native-uis-cheaper-than-terminal-interfaces-01M0JHXBD6WQPC975675X5VH4M Modelwire en 2026-08-21T16:07:32.000Z Coding agents make native UIs cheaper than terminal interfaces https://themodelwire.com/article/us-pressures-allies-to-abandon-chinese-ai-partnerships-01M0JEFJTJMXSGWAPHH1ZPE8C4 Modelwire en 2026-08-21T15:18:29.000Z US pressures allies to abandon Chinese AI partnerships https://themodelwire.com/article/matt-webb-learns-quaternions-via-chatgpt-tutoring-not-code-generation-01M0JEFFWPE1B21PSK2ENKQGPG Modelwire en 2026-08-21T15:06:26.000Z Matt Webb learns quaternions via ChatGPT tutoring, not code generation https://themodelwire.com/article/agentic-ai-tackles-semiconductor-yield-diagnosis-across-data-silos-01M0JDM2MWXWQH7VPM7APQYXT7 Modelwire en 2026-08-21T14:32:37.000Z Agentic AI tackles semiconductor yield diagnosis across data silos https://themodelwire.com/article/enterprise-ai-agents-outpace-organizational-readiness-for-deployment-01M0JCRM02W7M8ZMRDJJRHT9C2 Modelwire en 2026-08-21T14:28:42.000Z Enterprise AI agents outpace organizational readiness for deployment https://themodelwire.com/article/starcloud-secures-250m-to-build-space-based-ai-compute-as-launch-windows-narrow-01M0JA663F0PY72D6CSRMGCRQX Modelwire en 2026-08-21T14:00:00.000Z Starcloud secures $250M to build space-based AI compute as launch windows narrow https://themodelwire.com/article/doj-antitrust-probe-targets-a16z-board-conflicts-in-ai-data-infrastructure-01M0JH1XCMZXWE90BYCCACM5WV Modelwire en 2026-08-21T14:00:00.000Z DOJ antitrust probe targets a16z board conflicts in AI data infrastructure https://themodelwire.com/article/youtube-creators-face-backlash-for-sponsored-higgsfield-ai-video-demos-01M0J9APMP93VHEMMDAHA87XT8 Modelwire en 2026-08-21T13:37:52.000Z YouTube creators face backlash for sponsored Higgsfield AI video demos https://themodelwire.com/article/deepmind-embeds-game-ai-into-studio-production-pipelines-01M0J3ACC6JMSKBG2121BZNTWH Modelwire en 2026-08-21T11:59:48.000Z DeepMind embeds game AI into studio production pipelines https://themodelwire.com/article/waymo-develops-custom-chip-to-reduce-nvidia-dependency-in-robotaxi-fleet-01M0J0R4G4DNA53R6T4140YS1X Modelwire en 2026-08-21T11:04:09.000Z Waymo develops custom chip to reduce Nvidia dependency in robotaxi fleet https://themodelwire.com/article/meta-becomes-major-customer-for-microsofts-ai-infrastructure-01M0HZWNC0FA5EBJ21D5WAYE2Q Modelwire en 2026-08-21T10:54:41.000Z Meta becomes major customer for Microsoft's AI infrastructure https://themodelwire.com/article/anthropic-grants-enterprises-direct-control-over-conversation-data-01M0HWESPBAGQ2Z1FGJWY9M9DM Modelwire en 2026-08-21T09:59:51.000Z Anthropic grants enterprises direct control over conversation data https://themodelwire.com/article/nvidia-acquires-poolside-to-own-model-development-end-to-end-01M0HQA0785WFVHE81P0CPTV47 Modelwire en 2026-08-21T08:27:23.000Z Nvidia acquires Poolside to own model development end-to-end https://themodelwire.com/article/openai-reclaims-enterprise-api-lead-with-gpt-5-6-sol-launch-01M0HQA07J5V762R3RR7791301 Modelwire en 2026-08-21T08:26:05.000Z OpenAI reclaims enterprise API lead with GPT-5.6 Sol launch https://themodelwire.com/article/openai-adds-native-transparent-backgrounds-to-gpt-image-2-api-01M0HNK2JEPXFM20GHATSP92Z6 Modelwire en 2026-08-21T07:53:40.000Z OpenAI adds native transparent backgrounds to GPT-Image-2 API https://themodelwire.com/article/greater-manchester-rejects-uk-government-palantir-health-contract-01M0HFJRKGF521CR7WE8Y35V5D Modelwire en 2026-08-21T06:00:00.000Z Greater Manchester rejects UK government Palantir health contract https://themodelwire.com/article/micro1-hits-500m-run-rate-as-ai-training-data-becomes-strategic-bottleneck-01M0GTZHNV6PKBC9JWP52YKWDF Modelwire en 2026-08-21T00:13:44.000Z Micro1 hits $500M run rate as AI training data becomes strategic bottleneck https://themodelwire.com/article/platformer-trains-ai-agent-on-editors-30-000-copyedits-01M0GT449W8BZJQ7R3G05QN5X8 Modelwire en 2026-08-21T00:00:50.000Z Platformer trains AI agent on editor's 30,000 copyedits https://themodelwire.com/article/chatgpt-deploys-site-operator-at-scale-accelerating-generative-search-optimizati-01M0GTZH366M4QJKGHTTN7KC1N Modelwire en 2026-08-20T23:57:32.000Z ChatGPT deploys site operator at scale, accelerating generative search optimization https://themodelwire.com/article/enterprise-ai-spending-shows-no-loyalty-as-openai-gains-on-anthropic-01M0GNTREPMXFHR55WWTHG0CG8 Modelwire en 2026-08-20T22:36:37.000Z Enterprise AI spending shows no loyalty as OpenAI gains on Anthropic https://themodelwire.com/article/chatgpt-gains-native-integration-in-apple-messages-for-automated-text-compositio-01M0GM3TD3TKYQRW0VWSGH4V1Q Modelwire en 2026-08-20T22:09:51.000Z ChatGPT gains native integration in Apple Messages for automated text composition https://themodelwire.com/article/google-embeds-conversational-ai-into-discover-feed-personalization-01M0GK8CF98E3XWRJVAJ6AGF6B Modelwire en 2026-08-20T21:50:22.000Z Google embeds conversational AI into Discover feed personalization https://themodelwire.com/article/adobe-integrates-gemini-flash-and-launches-audio-generation-in-firefly-01M0GANRX3F78XA6AGBGQF0GS1 Modelwire en 2026-08-20T19:29:34.000Z Adobe integrates Gemini Flash and launches audio generation in Firefly https://themodelwire.com/article/google-adds-publisher-preference-controls-to-counter-ai-search-traffic-drain-01M0G9T7J0YAHBNHEST1REKJRN Modelwire en 2026-08-20T19:18:21.000Z Google adds publisher preference controls to counter AI search traffic drain https://themodelwire.com/article/tech-leaders-ignore-public-ai-concerns-while-doubling-down-on-promotion-01M0G8YSQ9JRBW1S65WQ11GEZE Modelwire en 2026-08-20T18:49:07.000Z Tech leaders ignore public AI concerns while doubling down on promotion https://themodelwire.com/article/new-benchmark-reveals-unlearning-methods-miss-harmful-concept-removal-01M0H1VB682C5CPJV6BCD62RF7 Modelwire en 2026-08-20T17:59:57.000Z New benchmark reveals unlearning methods miss harmful concept removal https://themodelwire.com/article/new-benchmark-tests-whether-ai-agents-can-design-better-training-algorithms-01M0H1VBAEVDSSN5TDCAKT4GQH Modelwire en 2026-08-20T17:56:59.000Z New benchmark tests whether AI agents can design better training algorithms https://themodelwire.com/article/safety-guardrails-not-capability-make-llm-text-detectable-01M0G4NGY4KR9BB2JCGNE06C1A Modelwire en 2026-08-20T17:36:51.000Z Safety guardrails, not capability, make LLM text detectable https://themodelwire.com/article/self-training-benchmarks-hide-systematic-measurement-artifacts-study-finds-01M0H1VBB8HB133G4CHW15V00J Modelwire en 2026-08-20T17:30:14.000Z Self-training benchmarks hide systematic measurement artifacts, study finds https://themodelwire.com/article/one-third-of-new-webpages-show-ai-authorship-since-chatgpt-launch-01M0G2YGNVAZPNZHD7B75GQQNZ Modelwire en 2026-08-20T17:18:58.000Z One-third of new webpages show AI authorship since ChatGPT launch https://themodelwire.com/article/subtask-level-skills-outperform-task-level-in-llm-agent-transfer-01M0H1VBG5ZFB8XMPNBC14YTXM Modelwire en 2026-08-20T17:12:08.000Z Subtask-level skills outperform task-level in LLM agent transfer https://themodelwire.com/article/hugging-face-lfm2-5-dspark-cuts-inference-latency-by-3-2x-01M0G22Z6498E91NE3508V8GAG Modelwire en 2026-08-20T16:52:57.000Z Hugging Face LFM2.5-DSpark cuts inference latency by 3.2x https://themodelwire.com/article/ramp-builds-llm-routing-layer-to-let-enterprises-swap-models-via-api-01M0G17M9QY7ES0J68H3PXW78H Modelwire en 2026-08-20T16:46:00.000Z Ramp builds LLM routing layer to let enterprises swap models via API https://themodelwire.com/article/cpu-first-architecture-beats-larger-models-without-the-cache-overhead-01M0H1VBHADV2CD92VH0DAA5ZJ Modelwire en 2026-08-20T16:09:43.000Z CPU-first architecture beats larger models without the cache overhead https://themodelwire.com/article/meta-expands-pocket-ai-game-creation-app-to-u-s-after-brazil-pilot-01M0FZGMG3WB1PVVQAAQQK7RX6 Modelwire en 2026-08-20T16:07:26.000Z Meta expands Pocket AI game-creation app to U.S. after Brazil pilot https://themodelwire.com/article/memory-augmented-llms-fail-when-past-context-corrupts-reasoning-01M0H1VBK6KS0GPMVKEJFRG5W0 Modelwire en 2026-08-20T16:00:17.000Z Memory-augmented LLMs fail when past context corrupts reasoning https://themodelwire.com/article/leadership-shift-at-openai-amid-legal-battles-and-ipo-preparation-01M0FYN72B95PSPCKQT0DB50DE Modelwire en 2026-08-20T15:45:55.000Z Leadership shift at OpenAI amid legal battles and IPO preparation https://themodelwire.com/article/mit-challenges-consciousness-framing-in-ai-safety-debate-01M0FYN57R97QSM54MC1SGPYBM Modelwire en 2026-08-20T15:42:39.000Z MIT challenges consciousness framing in AI safety debate https://themodelwire.com/article/contrastive-models-decode-silent-reading-from-consumer-eeg-hardware-01M0H2PT3PSJV863HQDKY1SFJ6 Modelwire en 2026-08-20T15:41:05.000Z Contrastive models decode silent reading from consumer EEG hardware https://themodelwire.com/article/relation-mechanism-outperforms-attention-across-model-scales-with-3-6x-speedup-01M0H2PT5H3ZSH31509KQZHYHB Modelwire en 2026-08-20T15:27:25.000Z Relation mechanism outperforms attention across model scales with 3.6x speedup https://themodelwire.com/article/autoformalization-emerges-as-core-bottleneck-in-llm-theorem-proving-01M0H1VBMHYZCJ5786175ZRAAM Modelwire en 2026-08-20T15:13:41.000Z Autoformalization emerges as core bottleneck in LLM theorem proving https://themodelwire.com/article/llms-show-systematic-bias-when-text-and-numbers-conflict-01M0H1VBNTYG3W5HBHMJ70XVBY Modelwire en 2026-08-20T14:48:30.000Z LLMs show systematic bias when text and numbers conflict https://themodelwire.com/article/chinese-models-close-gap-with-us-frontier-labs-eroding-capability-moat-01M0FRMZP6SCZA5MGMSDWMPRRR Modelwire en 2026-08-20T14:08:14.000Z Chinese models close gap with US frontier labs, eroding capability moat https://themodelwire.com/article/openais-math-breakthroughs-trigger-existential-debate-among-mathematicians-01M0FQSETHB360V0HKYDHT9KY1 Modelwire en 2026-08-20T14:00:00.000Z OpenAI's math breakthroughs trigger existential debate among mathematicians https://themodelwire.com/article/hyperparameter-transfer-method-cuts-tuning-costs-for-trillion-token-moe-training-01M0H1VBSHBJ3CVGFQE78C1QQF Modelwire en 2026-08-20T13:57:43.000Z Hyperparameter transfer method cuts tuning costs for trillion-token MoE training