Amazon duda de Anthropic, BaFin regula la IA bancaria y la UE prohíbe deepfakes

AI Radar ES - Update 2026-07-19

The latest AI news updates for ES. The candidate stories provided for the briefing dated July 19, 2026, have been evaluated against our standards for source authority, corroboration, and recency. All submitted candidate stories have been **DROPPED**. **Reasoning:** 1. **Source Authority:** None of the provided stories rely on our verified allowlist of reputable outlets (e.g., *Reuters, Bloomberg, TechCrunch, The Verge, OpenAI, Nvidia, etc.*). 2. **Banned/Unverified Sources:** The candidates rely exclusively on banned or low-...

AI Radar ES - Update 2026-07-18

The latest AI news updates for ES. ### **FINAL VETTED BRIEFING** #### **1. Nvidia Partners with Japan for "FRONTia" National AI Infrastructure** * **FACTS:** Nvidia is partnering with Noetra Corp. (a consortium including SoftBank, Sony, NEC, and Honda) to build a 140MW AI factory in Japan. The facility will utilize 27,500 Rubin GPUs and 13,750 Vera CPUs based on the Vera Rubin NVL72 rack architecture. This infrastructure supports the government-backed "FRONTia" project, managed by the Ministry of Economy, Trade and Industry (M...

OpenAI revela GPT-Red | Nvidia lanza IA física en Japón | Nace Kimi K3

¿Puede la IA auditarse a sí misma? OpenAI presenta su nuevo modelo interno para buscar vulnerabilidades, mientras Japón apuesta por la robótica soberana. En este episodio: • Nvidia y el consorcio japonés Noetra anuncian "FRONTia", una infraestructura nacional de 140MW para IA física y gemelos digitales. • OpenAI devela "GPT-Red", un modelo de red-teaming automatizado para mejorar la seguridad. • Meta endurece la protección en Meta AI alertando a los padres sobre riesgos en adolescentes. • Demis Hassabis (Google DeepMind) propone un regulador independiente para la IA de frontera. • Moonshot AI sorprende con Kimi K3, un modelo de 2.8 billones de parámetros. Sources: • Tom's Hardware: https://www.tomshardware.com [https://www.tomshardware.com] • Data Centre Magazine: https://datacentremagazine.com [https://datacentremagazine.com] • OpenAI: https://openai.com [https://openai.com] • The Hacker News: https://thehackernews.com [https://thehackernews.com] • Meta: https://about.fb.com/news/2026/07/alerting-parents-if-teens-show-signs-of-distress-in-conversations-with-meta-ai [https://about.fb.com/news/2026/07/alerting-parents-if-teens-show-signs-of-distress-in-conversations-with-meta-ai] • Mashable: https://mashable.com [https://mashable.com] • Axios: https://www.axios.com/2026/07/16/google-deepmind-biosecurity-ai-regulation [https://www.axios.com/2026/07/16/google-deepmind-biosecurity-ai-regulation] • PYMNTS: https://www.pymnts.com [https://www.pymnts.com] • VentureBeat: https://venturebeat.com [https://venturebeat.com] • Kimi AI: https://kimi.ai [https://kimi.ai]

Anthropic lanza Ode ($1.5B) | Meta demandada por despidos con IA | Huawei Atlas 950

Meta enfrenta demanda por despidos IA. Anthropic lanza **Ode** ($1.5B), llega **Inkling** y Huawei revela Atlas 950. Sources: • TC: https://techcrunch.com/2026/07/15/anthropic-blackstone-launch-1-5b-ode-enterprise-services-firm/ [https://techcrunch.com/2026/07/15/anthropic-blackstone-launch-1-5b-ode-enterprise-services-firm/] • SA: https://seekingalpha.com/news/4414605-blackstone-anthropic-hellman-friedmans-ai-firm-launches-under-ode-with-anthropic-name [https://seekingalpha.com/news/4414605-blackstone-anthropic-hellman-friedmans-ai-firm-launches-under-ode-with-anthropic-name] • TM: https://thinkingmachines.ai/news/introducing-inkling [https://thinkingmachines.ai/news/introducing-inkling] • Baseten: https://baseten.co/blog/inkling-day-0-support/ [https://baseten.co/blog/inkling-day-0-support/] • SCMP: https://www.scmp.com/tech/tech-war/article/3273394/huaweis-new-computing-cluster-worlds-first-ai-agent-phone-debut-china-ai-summit [https://www.scmp.com/tech/tech-war/article/3273394/huaweis-new-computing-cluster-worlds-first-ai-agent-phone-debut-china-ai-summit] • DT: https://www.digitimes.com/news/a20260715VL201/huawei-waic-atlas-950-superpod.html [https://www.digitimes.com/news/a20260715VL201/huawei-waic-atlas-950-superpod.html] • Gizmodo: https://gizmodo.com/meta-sued-for-allegedly-using-discriminatory-ai-in-layoff-decisions-2000574041 [https://gizmodo.com/meta-sued-for-allegedly-using-discriminatory-ai-in-layoff-decisions-2000574041] • MJ: https://www.motherjones.com/politics/2026/07/lawsuit-meta-used-ai-for-discriminatory-layoffs/ [https://www.motherjones.com/politics/2026/07/lawsuit-meta-used-ai-for-discriminatory-layoffs/] • Wired: https://www.wired.com [https://www.wired.com]

Apple demanda a OpenAI: Guerra por hardware. TSMC rompe récords y Netflix GenPage

¿Por qué Apple está demandando a OpenAI? La guerra por el talento y el hardware de IA acaba de estallar. En este episodio analizamos la escalada judicial entre **Apple** y **OpenAI**, acusada de robar diseños y a 400 exempleados. Además, examinamos cómo la insaciable demanda de chips llevó a **TSMC** a ingresos récord de $39.6B, confirmando la solidez del sector. También desglosamos **GenPage** de **Netflix**, un modelo generativo que construye la interfaz de usuario en tiempo real. Por último, los rumores sobre la posible cesión de un 5% del capital de OpenAI al gobierno de EE. UU. Sources: • MacRumors: https://www.macrumors.com [https://www.macrumors.com] • The Guardian: https://www.theguardian.com [https://www.theguardian.com] • Reuters: https://www.reuters.com [https://www.reuters.com] • Bloomberg: https://www.bloomberg.com [https://www.bloomberg.com] • Netflix Tech Blog: https://netflixtechblog.com/genpage-towards-end-to-end-generative-homepage-construction-at-netflix-77146fba8a08 [https://netflixtechblog.com/genpage-towards-end-to-end-generative-homepage-construction-at-netflix-77146fba8a08] • Arxiv: https://arxiv.org [https://arxiv.org] • Financial Times: https://www.ft.com [https://www.ft.com] • Axios: https://www.axios.com [https://www.axios.com]