https://themodelwire.com/article/anthropic-frames-30-trillion-market-ahead-of-ipo-filing-01M0YRECBFQWZNJM2QS552YR86 Modelwire en 2026-08-26T10:02:39.000Z Anthropic frames $30 trillion market ahead of IPO filing https://themodelwire.com/article/apple-and-openai-push-ai-compute-to-devices-sidestepping-nvidia-01M0YRE9YNS9Q8PGW62CMNAV2A Modelwire en 2026-08-26T10:00:00.000Z Apple and OpenAI push AI compute to devices, sidestepping Nvidia https://themodelwire.com/article/loveholidays-ships-73-more-with-flat-engineering-team-using-codex-01M0YNVZ03XHZP5JJNSMH5C9VJ Modelwire en 2026-08-26T09:06:31.000Z Loveholidays ships 73% more with flat engineering team using Codex https://themodelwire.com/article/current-ai-models-fail-standard-reasoning-benchmarks-01M0YNVWR7VQGR4MKJA4QERE78 Modelwire en 2026-08-26T09:00:00.000Z Current AI models fail standard reasoning benchmarks https://themodelwire.com/article/ai-generated-million-line-codebase-now-powers-millions-of-developer-machines-01M0YJE1GV59M86WHG4V4S8ZGJ Modelwire en 2026-08-26T08:07:55.000Z AI-generated million-line codebase now powers millions of developer machines https://themodelwire.com/article/gates-declares-ai-has-crossed-danger-thresholds-pivots-to-response-01M0YF05HZ11JCVCH2V2WRBCA3 Modelwire en 2026-08-26T07:01:00.000Z Gates declares AI has crossed danger thresholds, pivots to response https://themodelwire.com/article/generalist-robotics-startup-hits-3b-valuation-on-embodied-ai-momentum-01M0XRP2JWCKQJ12CYDNTGK59R Modelwire en 2026-08-26T00:40:59.000Z Generalist robotics startup hits $3B valuation on embodied AI momentum https://themodelwire.com/article/openai-restructures-infrastructure-leadership-as-data-center-exec-departs-01M0XPZ5A3XS1EMR27QVTD1HAN Modelwire en 2026-08-26T00:06:20.000Z OpenAI restructures infrastructure leadership as data center exec departs https://themodelwire.com/article/loveholidays-deploys-codex-to-scale-development-beyond-engineers-01M0YRE9CHYMKCX7D1KP54WAS9 Modelwire en 2026-08-26T00:00:00.000Z Loveholidays deploys Codex to scale development beyond engineers https://themodelwire.com/article/openai-embeds-workspace-administration-into-chatgpt-work-01M0XA36YSG9RK1SS29V0M181Q Modelwire en 2026-08-25T20:17:22.000Z OpenAI embeds workspace administration into ChatGPT Work https://themodelwire.com/article/openai-embeds-agent-tools-directly-into-web-interfaces-via-webmcp-01M0XA36ZXRBYEFBHD4CW7Z5E5 Modelwire en 2026-08-25T20:15:30.000Z OpenAI embeds agent tools directly into web interfaces via WebMCP https://themodelwire.com/article/google-launches-financial-and-legal-ai-agents-amid-competitive-pressure-01M0X7GTJ1NRWVJBEAZP1KWBDC Modelwire en 2026-08-25T19:38:38.000Z Google launches financial and legal AI agents amid competitive pressure https://themodelwire.com/article/stability-ai-secures-76-million-as-open-weight-image-generation-attracts-sustain-01M0X4YC4E9JQ12RKF0C8Y1A3X Modelwire en 2026-08-25T19:03:47.000Z Stability AI secures $76 million as open-weight image generation attracts sustained capital https://themodelwire.com/article/openai-disrupts-russian-disinformation-operation-using-chatgpt-at-scale-01M0X4YE58PP86G3R9VFV2KJN1 Modelwire en 2026-08-25T18:54:56.000Z OpenAI disrupts Russian disinformation operation using ChatGPT at scale https://themodelwire.com/article/five-voice-ai-builders-expose-production-gaps-between-demos-and-live-systems-01M0X4YDBSSPE3BRMDRWNVSCXT Modelwire en 2026-08-25T18:45:32.000Z Five voice AI builders expose production gaps between demos and live systems https://themodelwire.com/article/state-backed-operation-floods-ai-training-data-with-synthetic-content-01M0X42Y8CYQ7J8MRY5MTBQW4K Modelwire en 2026-08-25T18:35:26.000Z State-backed operation floods AI training data with synthetic content https://themodelwire.com/article/ice-contracts-voter-data-analysis-to-identify-immigration-fraud-01M0X2C00WNXP41A0BFVSH98WV Modelwire en 2026-08-25T18:16:34.000Z ICE contracts voter data analysis to identify immigration fraud https://themodelwire.com/article/openais-custom-inference-chip-outperforms-nvidias-latest-generation-01M0X1GMAR4KGWK4HHWCD3XEV0 Modelwire en 2026-08-25T18:00:50.000Z OpenAI's custom inference chip outperforms Nvidia's latest generation https://themodelwire.com/article/fid-metric-blindness-lets-garbage-images-score-like-real-data-01M0XXTXM1NAJBPRM6E70QCCKY Modelwire en 2026-08-25T17:58:46.000Z FID metric blindness lets garbage images score like real data https://themodelwire.com/article/recuris-decouples-memory-to-improve-long-horizon-agent-reasoning-01M0XYPD4FMBC2GKWE6DX3W1NZ Modelwire en 2026-08-25T17:56:35.000Z Recuris decouples memory to improve long-horizon agent reasoning https://themodelwire.com/article/anthropic-unifies-claude-memory-across-chat-and-cowork-01M0X1GH9R3H88ZHBTB4T4GPK5 Modelwire en 2026-08-25T17:50:33.000Z Anthropic unifies Claude memory across chat and Cowork https://themodelwire.com/article/browserforge-generates-web-agent-training-data-across-parallel-sandboxes-01M0XYPD66SSM4BFND0P66JQP9 Modelwire en 2026-08-25T17:35:42.000Z BrowserForge generates web agent training data across parallel sandboxes https://themodelwire.com/article/laion-releases-10-million-hour-open-video-dataset-for-multimodal-training-01M0XXTXNSCZ4JDN76QBW3G1FE Modelwire en 2026-08-25T17:34:19.000Z LAION releases 10-million-hour open video dataset for multimodal training https://themodelwire.com/article/llms-retrieve-financial-data-but-ignore-it-in-investment-decisions-01M0XYPD7BFPY923Z1KM7YH8VF Modelwire en 2026-08-25T17:31:25.000Z LLMs retrieve financial data but ignore it in investment decisions https://themodelwire.com/article/openai-removes-chat-limits-from-gpt-5-6-luna-rollout-01M0WZSM8AGMPMXNQ3108JF3AJ Modelwire en 2026-08-25T17:21:58.000Z OpenAI removes chat limits from GPT-5.6 Luna rollout https://themodelwire.com/article/effective-learning-rate-ratio-predicts-loss-dynamics-across-model-scales-01M0XXTXPMQW41QMASW6BNK407 Modelwire en 2026-08-25T16:57:29.000Z Effective learning rate ratio predicts loss dynamics across model scales https://themodelwire.com/article/structured-citation-graphs-outperform-open-ended-research-agents-by-3x-01M0XYPD8RSDTMMN78Z7YQZEY1 Modelwire en 2026-08-25T16:51:26.000Z Structured citation graphs outperform open-ended research agents by 3x https://themodelwire.com/article/linear-probes-outperform-ai-text-detectors-with-minimal-training-data-01M0XYPD9E981FJD1ED9HDFSRM Modelwire en 2026-08-25T16:21:24.000Z Linear probes outperform AI text detectors with minimal training data https://themodelwire.com/article/google-embeds-gemini-into-legal-workflows-via-enterprise-connectors-01M0WWBTJEVVJJWTHBC7AZF6EA Modelwire en 2026-08-25T16:21:09.000Z Google embeds Gemini into legal workflows via enterprise connectors https://themodelwire.com/article/bayesian-framework-exposes-hidden-behavioral-differences-in-rag-systems-01M0XYPDCZ31KZ2BYN4QSNAJ5J Modelwire en 2026-08-25T15:54:30.000Z Bayesian framework exposes hidden behavioral differences in RAG systems https://themodelwire.com/article/recursive-framework-removes-depth-ceiling-for-self-improving-llm-agents-01M0XYPDF01D2EH5YRMTQMAC79 Modelwire en 2026-08-25T15:44:25.000Z Recursive framework removes depth ceiling for self-improving LLM agents https://themodelwire.com/article/first-multilingual-spoken-hallucination-benchmark-spans-three-languages-and-audi-01M0XYPDGZJ84PF56115JBWMH9 Modelwire en 2026-08-25T15:24:48.000Z First multilingual spoken hallucination benchmark spans three languages and audio formats https://themodelwire.com/article/llms-show-near-zero-calibration-on-hidden-information-decisions-01M0XXTXWQNR8E55KPJ9HA1GM9 Modelwire en 2026-08-25T15:18:18.000Z LLMs show near-zero calibration on hidden information decisions https://themodelwire.com/article/foundation-model-trained-on-1-67m-cancer-patients-improves-oncology-prognosis-01M0XXTXWZXC4YSHNZV47014B9 Modelwire en 2026-08-25T15:17:12.000Z Foundation model trained on 1.67M cancer patients improves oncology prognosis https://themodelwire.com/article/ibm-publishes-granite-4-2-training-methodology-opening-enterprise-llm-architectu-01M0WR2AC96BDGD58Z2BR5FFFE Modelwire en 2026-08-25T15:14:14.000Z IBM publishes Granite 4.2 training methodology, opening enterprise LLM architecture https://themodelwire.com/article/maia-200-accelerator-prioritizes-data-movement-over-thread-parallelism-01M0XXTXXH7CYSQ55WP0HXSY1M Modelwire en 2026-08-25T15:05:40.000Z Maia 200 accelerator prioritizes data movement over thread parallelism https://themodelwire.com/article/hidden-inference-steering-reshapes-llm-outputs-without-disclosure-01M0XYPDN8TGJQ848Z27FGECHK Modelwire en 2026-08-25T15:05:19.000Z Hidden inference steering reshapes LLM outputs without disclosure https://themodelwire.com/article/ibm-releases-granite-speech-5-0-turbo-for-low-latency-transcription-01M0WQ6VDWD4HQXVQS1SKK7ZAK Modelwire en 2026-08-25T15:01:00.000Z IBM releases Granite Speech 5.0 Turbo for low-latency transcription https://themodelwire.com/article/reddit-sentiment-tracks-model-releases-not-just-capabilities-01M0XYPDPG81XMR1E8JQJN67QY Modelwire en 2026-08-25T15:00:44.000Z Reddit sentiment tracks model releases, not just capabilities https://themodelwire.com/article/air-traffic-control-study-exposes-semantic-safety-gap-in-language-models-01M0XYPDQN5NQGJ9N89GDBPVTP Modelwire en 2026-08-25T14:39:02.000Z Air traffic control study exposes semantic-safety gap in language models https://themodelwire.com/article/xpeng-robotics-unit-reaches-6-3b-valuation-on-humanoid-expansion-01M0WR2DYGTDR1AH26JJFSH5F8 Modelwire en 2026-08-25T14:36:52.000Z XPeng robotics unit reaches $6.3B valuation on humanoid expansion https://themodelwire.com/article/openais-jalapeno-chip-beats-inference-benchmarks-on-throughput-and-efficiency-01M0WNFZD2SAN3F77C0J0QA05V Modelwire en 2026-08-25T14:22:04.000Z OpenAI's Jalapeño chip beats inference benchmarks on throughput and efficiency https://themodelwire.com/article/mistral-builds-middle-east-sovereign-ai-foothold-with-saudi-partner-01M0WNG1AGK1Y3Y31NYJJ5YHM8 Modelwire en 2026-08-25T14:12:00.000Z Mistral builds Middle East sovereign AI foothold with Saudi partner https://themodelwire.com/article/openai-launches-jalapeno-chip-to-cut-ai-inference-latency-01M0WKS2P015KZB48THD9GMTH3 Modelwire en 2026-08-25T14:00:00.000Z OpenAI launches Jalapeño chip to cut AI inference latency https://themodelwire.com/article/meta-launches-paid-ai-agent-hatch-with-new-watermelon-model-in-october-01M0WJXN57A5Y80JYGVJH3PR5K Modelwire en 2026-08-25T13:43:40.000Z Meta launches paid AI agent Hatch with new Watermelon model in October https://themodelwire.com/article/keenable-launches-agent-first-search-index-with-26m-seed-01M0WJ23GA1GTMYQRHXV0NMN9J Modelwire en 2026-08-25T13:00:00.000Z Keenable launches agent-first search index with $26M seed https://themodelwire.com/article/predicting-post-training-neuron-shifts-improves-parameter-efficient-fine-tuning-01M0XYPDT5KER25EW0ZKTF9N2Y Modelwire en 2026-08-25T12:28:55.000Z Predicting post-training neuron shifts improves parameter-efficient fine-tuning https://themodelwire.com/article/nvidias-groq-3-lpx-claims-speed-lead-but-efficiency-math-favors-cerebras-01M0WEMAT79T0JW2TCD44C2T3R Modelwire en 2026-08-25T12:27:57.000Z Nvidia's Groq 3 LPX claims speed lead, but efficiency math favors Cerebras https://themodelwire.com/article/ukraine-grants-uk-first-access-to-five-million-combat-images-for-weapons-ai-trai-01M0WDRW3JCVN1CC1BZX8Z40QG Modelwire en 2026-08-25T12:16:41.000Z Ukraine grants UK first access to five million combat images for weapons AI training https://themodelwire.com/article/transformers-solve-conflicts-via-unstable-heuristics-not-robust-reasoning-01M0XYPDV6HKVE87335345XXRK Modelwire en 2026-08-25T12:09:19.000Z Transformers solve conflicts via unstable heuristics, not robust reasoning