https://aiexpert.news/en/article/dropbox-uses-mcp-to-automate-security-reviews-in-code ai|expert en 2026-07-31T18:14:09.311Z Dropbox Automates Security Reviews Using MCP Retrieval https://aiexpert.news/pt/article/dropbox-uses-mcp-to-automate-security-reviews-in-code ai|expert pt-BR 2026-07-31T18:14:09.311Z Dropbox Automatiza Revisões de Segurança Usando Recuperação MCP https://aiexpert.news/es/article/dropbox-uses-mcp-to-automate-security-reviews-in-code ai|expert es 2026-07-31T18:14:09.311Z Dropbox Automatiza Revisiones de Seguridad Usando Recuperación MCP https://aiexpert.news/en/article/how-cloudflare-runs-cdnjs-on-its-own-developer-platform-at-global-scale ai|expert en 2026-07-31T17:42:09.588Z Cloudflare Consolidates cdnjs Pipeline on Its Own Platform https://aiexpert.news/pt/article/how-cloudflare-runs-cdnjs-on-its-own-developer-platform-at-global-scale ai|expert pt-BR 2026-07-31T17:42:09.588Z Cloudflare Consolida o Pipeline cdnjs em Sua Própria Plataforma https://aiexpert.news/es/article/how-cloudflare-runs-cdnjs-on-its-own-developer-platform-at-global-scale ai|expert es 2026-07-31T17:42:09.588Z Cloudflare Consolida el Pipeline de cdnjs en Su Propia Plataforma https://aiexpert.news/en/article/langsmith-llm-gateway-enforces-runtime-controls-for-agentic-inference-loops ai|expert en 2026-07-31T17:10:09.696Z LangChain embeds governance into LangSmith to stop runaway agentic calls https://aiexpert.news/pt/article/langsmith-llm-gateway-enforces-runtime-controls-for-agentic-inference-loops ai|expert pt-BR 2026-07-31T17:10:09.696Z LangChain integra governança no LangSmith para parar chamadas de agentes descontroladas https://aiexpert.news/es/article/langsmith-llm-gateway-enforces-runtime-controls-for-agentic-inference-loops ai|expert es 2026-07-31T17:10:09.696Z LangChain integra gobernanza en LangSmith para detener llamadas de agentes descontroladas https://aiexpert.news/en/article/osreward-brings-standardized-grading-to-multi-platform-agent-evaluation ai|expert en 2026-07-31T16:38:09.413Z OSReward exposes systematic bias in VLM judges for agent evaluation https://aiexpert.news/pt/article/osreward-brings-standardized-grading-to-multi-platform-agent-evaluation ai|expert pt-BR 2026-07-31T16:38:09.413Z OSReward expõe viés sistemático em juízes VLM para avaliação de agentes https://aiexpert.news/es/article/osreward-brings-standardized-grading-to-multi-platform-agent-evaluation ai|expert es 2026-07-31T16:38:09.413Z OSReward expone sesgo sistemático en jueces VLM para evaluación de agentes https://aiexpert.news/en/article/silicon-photonics-hits-a-hard-supply-wall30-below-customer-needs ai|expert en 2026-07-31T16:04:09.850Z Indium Phosphide Shortage Deepens as Nvidia Pumps $4 Billion Into Laser Makers https://aiexpert.news/pt/article/silicon-photonics-hits-a-hard-supply-wall30-below-customer-needs ai|expert pt-BR 2026-07-31T16:04:09.850Z Escassez de Fosfeto de Índio se Aprofunda Conforme Nvidia Investe $4 Bilhões em Fabricantes de Lasers https://aiexpert.news/es/article/silicon-photonics-hits-a-hard-supply-wall30-below-customer-needs ai|expert es 2026-07-31T16:04:09.850Z La Escasez de Fosfuro de Indio se Profundiza Mientras Nvidia Invierte $4 Mil Millones en Fabricantes de Láseres https://aiexpert.news/en/article/netflixs-llm-native-rec-engine-cuts-latency-while-preserving-personalization ai|expert en 2026-07-31T15:14:09.540Z Netflix Cuts Ranking Labels 40× by Switching to LLM-Native System https://aiexpert.news/pt/article/netflixs-llm-native-rec-engine-cuts-latency-while-preserving-personalization ai|expert pt-BR 2026-07-31T15:14:09.540Z Netflix Reduz Rótulos de Classificação em 40× ao Migrar para Sistema Nativo com LLM https://aiexpert.news/es/article/netflixs-llm-native-rec-engine-cuts-latency-while-preserving-personalization ai|expert es 2026-07-31T15:14:09.540Z Netflix Reduce Etiquetas de Clasificación 40× al Cambiar a Sistema Nativo con LLM https://aiexpert.news/en/article/denseon-with-lateon-open-source-dense-and-late-interaction-retrieval-models-at-s ai|expert en 2026-07-31T14:40:09.629Z LightOn's LateOn outperforms 4× larger models while open-source https://aiexpert.news/pt/article/denseon-with-lateon-open-source-dense-and-late-interaction-retrieval-models-at-s ai|expert pt-BR 2026-07-31T14:40:09.629Z LateOn da LightOn supera modelos 4× maiores enquanto é código aberto https://aiexpert.news/es/article/denseon-with-lateon-open-source-dense-and-late-interaction-retrieval-models-at-s ai|expert es 2026-07-31T14:40:09.629Z LateOn de LightOn supera modelos 4× más grandes siendo de código abierto https://aiexpert.news/en/article/the-social-cost-of-an-ai-teammate-how-ai-reshapes-human-human-communication ai|expert en 2026-07-31T14:08:09.555Z AI Teammates Dominate Talks, Cut Human-to-Human Exchange https://aiexpert.news/pt/article/the-social-cost-of-an-ai-teammate-how-ai-reshapes-human-human-communication ai|expert pt-BR 2026-07-31T14:08:09.555Z Agentes de IA Dominam Discussões, Reduzem Troca Humano-Humano https://aiexpert.news/es/article/the-social-cost-of-an-ai-teammate-how-ai-reshapes-human-human-communication ai|expert es 2026-07-31T14:08:09.555Z Agentes de IA Dominan Conversaciones, Reducen Intercambio Humano-Humano https://aiexpert.news/en/article/omegause-officeval-benchmarking-llm-agents-on-long-horizon-office-tasks-with-cos ai|expert en 2026-07-31T13:36:09.974Z First office benchmark to price agent work reveals productivity gap https://aiexpert.news/pt/article/omegause-officeval-benchmarking-llm-agents-on-long-horizon-office-tasks-with-cos ai|expert pt-BR 2026-07-31T13:36:09.974Z Primeiro benchmark de escritório para precificar trabalho de agentes revela lacuna de produtividade https://aiexpert.news/es/article/omegause-officeval-benchmarking-llm-agents-on-long-horizon-office-tasks-with-cos ai|expert es 2026-07-31T13:36:09.974Z Primer benchmark de oficina para preciar el trabajo de agentes revela brecha de productividad https://aiexpert.news/en/article/apex-accounting-real-world-benchmark-for-frontier-models-on-accounting-workflows ai|expert en 2026-07-30T03:58:09.643Z No frontier model exceeds 2.6% on eight-step accounting tasks https://aiexpert.news/pt/article/apex-accounting-real-world-benchmark-for-frontier-models-on-accounting-workflows ai|expert pt-BR 2026-07-30T03:58:09.643Z Nenhum modelo de fronteira excede 2.6% em tarefas contábeis de oito etapas https://aiexpert.news/es/article/apex-accounting-real-world-benchmark-for-frontier-models-on-accounting-workflows ai|expert es 2026-07-30T03:58:09.643Z Ningún modelo de frontera supera el 2.6% en tareas contables de ocho pasos https://aiexpert.news/en/article/metas-capacity-crunch-forces-capex-reallocation-between-products-and-platform ai|expert en 2026-07-30T03:26:09.903Z Meta Considers $10B Compute Lease to Anthropic as GPU Demand Swells https://aiexpert.news/pt/article/metas-capacity-crunch-forces-capex-reallocation-between-products-and-platform ai|expert pt-BR 2026-07-30T03:26:09.903Z Meta Considera Locação de Computação de $10B para Anthropic com Demanda de GPU em Expansão https://aiexpert.news/es/article/metas-capacity-crunch-forces-capex-reallocation-between-products-and-platform ai|expert es 2026-07-30T03:26:09.903Z Meta Considera Arrendar Capacidad de Computación de $10B a Anthropic mientras la Demanda de GPU se Dispara https://aiexpert.news/en/article/openais-3x-arc-agi-3-boost-two-settings-no-model-retrain ai|expert en 2026-07-30T01:52:09.665Z OpenAI's API tweak lifts GPT-5.6 Sol to 38.3% on ARC-AGI-3 https://aiexpert.news/pt/article/openais-3x-arc-agi-3-boost-two-settings-no-model-retrain ai|expert pt-BR 2026-07-30T01:52:09.665Z Ajuste na API da OpenAI eleva GPT-5.6 Sol para 38,3% no ARC-AGI-3 https://aiexpert.news/es/article/openais-3x-arc-agi-3-boost-two-settings-no-model-retrain ai|expert es 2026-07-30T01:52:09.665Z El ajuste de API de OpenAI eleva GPT-5.6 Sol a 38,3% en ARC-AGI-3 https://aiexpert.news/en/article/why-agent-benchmarks-miss-the-state-verification-problemand-why-it-matters-for-w ai|expert en 2026-07-30T01:20:09.600Z Frontier Agents Hit 65% on State Verification in Production Workflows https://aiexpert.news/pt/article/why-agent-benchmarks-miss-the-state-verification-problemand-why-it-matters-for-w ai|expert pt-BR 2026-07-30T01:20:09.600Z Agentes de Fronteira Atingem 65% em Verificação de Estado em Fluxos de Produção https://aiexpert.news/es/article/why-agent-benchmarks-miss-the-state-verification-problemand-why-it-matters-for-w ai|expert es 2026-07-30T01:20:09.600Z Agentes Frontera Alcanzan 65% en Verificación de Estado en Flujos de Producción https://aiexpert.news/en/article/deep-agents-v07-advancing-the-agent-framework-for-production-tool-orchestration ai|expert en 2026-07-30T00:00:56.669Z LangChain Deep Agents v0.7 Cuts Prompt Tokens 65% https://aiexpert.news/pt/article/deep-agents-v07-advancing-the-agent-framework-for-production-tool-orchestration ai|expert pt-BR 2026-07-30T00:00:56.669Z LangChain Deep Agents v0.7 Reduz Tokens de Prompt em 65% https://aiexpert.news/es/article/deep-agents-v07-advancing-the-agent-framework-for-production-tool-orchestration ai|expert es 2026-07-30T00:00:56.669Z LangChain Deep Agents v0.7 Reduce Tokens de Prompt en 65% https://aiexpert.news/en/article/white-house-ai-framework-deadline-imminent-what-code-changes-might-hit-architect ai|expert en 2026-07-29T20:40:09.479Z August 1 Deadline Triggers Federal AI Framework for Covered Models https://aiexpert.news/pt/article/white-house-ai-framework-deadline-imminent-what-code-changes-might-hit-architect ai|expert pt-BR 2026-07-29T20:40:09.479Z Prazo de 1º de Agosto Ativa Estrutura Federal de IA para Modelos Cobertos https://aiexpert.news/es/article/white-house-ai-framework-deadline-imminent-what-code-changes-might-hit-architect ai|expert es 2026-07-29T20:40:09.479Z Plazo del 1 de Agosto Activa Marco Federal de IA para Modelos Cubiertos https://aiexpert.news/en/article/dynamic-ai-demands-drive-memory-diversity-in-inference-stacks ai|expert en 2026-07-29T20:12:09.371Z Micron's SOCAMM2 Module Overhauls AI Memory Hierarchy https://aiexpert.news/pt/article/dynamic-ai-demands-drive-memory-diversity-in-inference-stacks ai|expert pt-BR 2026-07-29T20:12:09.371Z Módulo SOCAMM2 da Micron Reformula a Hierarquia de Memória para IA https://aiexpert.news/es/article/dynamic-ai-demands-drive-memory-diversity-in-inference-stacks ai|expert es 2026-07-29T20:12:09.371Z El Módulo SOCAMM2 de Micron Revoluciona la Jerarquía de Memoria para IA https://aiexpert.news/en/article/agents-for-production-lines-trusted-decisions-in-real-time ai|expert en 2026-07-29T19:40:09.731Z Databricks ProdLine CoPilot speeds manufacturing decisions in-shift https://aiexpert.news/pt/article/agents-for-production-lines-trusted-decisions-in-real-time ai|expert pt-BR 2026-07-29T19:40:09.731Z Databricks ProdLine CoPilot acelera decisões de manufatura em tempo real https://aiexpert.news/es/article/agents-for-production-lines-trusted-decisions-in-real-time ai|expert es 2026-07-29T19:40:09.731Z Databricks ProdLine CoPilot acelera decisiones de manufactura en tiempo real https://aiexpert.news/en/article/ai-labs-sign-pace-letter-hf-cyberattack-governance-and-supply-chain-security-und ai|expert en 2026-07-29T19:08:09.766Z OpenAI Model's 17,600 Hacking Actions Disrupt Governance Letter's Timing https://aiexpert.news/pt/article/ai-labs-sign-pace-letter-hf-cyberattack-governance-and-supply-chain-security-und ai|expert pt-BR 2026-07-29T19:08:09.766Z Ações de Hackers do Modelo OpenAI 17.600 Ações de Hackers Prejudicam Cronograma da Carta de Governança https://aiexpert.news/es/article/ai-labs-sign-pace-letter-hf-cyberattack-governance-and-supply-chain-security-und ai|expert es 2026-07-29T19:08:09.766Z Las 17.600 Acciones de Hackers del Modelo OpenAI Interrumpen el Calendario de la Carta de Gobernanza