Transitioning to a hierarchical architecture utilizing HBM, Host DRAM, and NVMe SSD is essential to resolve AI inference memory bottlenecks caused by the proliferation of long contexts and RAG.
ChatGPT Ads, which achieved $1 billion in annualized revenue within 200 days of launch, has expanded its service to over 60 countries by adding seven Asian markets, including Indonesia and Taiwan.
Kakao introduced a personalized ranking model combining GDCN and MoE, improving the valid play rate by 20%.
Woowa Brothers will host the technical conference WOOWACON 2026 in Seoul on October 28, with registration open from September 22 to October 13.
Airbnb has expanded access to frontier models such as GPT-6 Astra through an agreement with OpenAI, resulting in an approximately 80% year-over-year increase in its development team's feature release speed.
Narae Energy Services successfully validated a proof-of-concept for automating power plant equipment diagnostics using Strands Agents and Bedrock AgentCore.
The MiMo-V2.6-Pro released by Xiaomi recorded higher intelligence scores compared to similar-sized open-weight models and features competitive pricing, though its output speed was found to be slower than average.
The rootless sandbox tool 'Drop' has been released, blocking dangerous command execution by AI coding agents at the OS level and supporting gVisor.
With approximately 1,500 attendees, the event presented next-generation AIDC technologies such as Liquid Cooling and a vision for public cloud expansion.
Underdog has released 'Husky', a model-specific inference engine that is faster than MLX across all measured metrics, achieving up to 4.5x performance gains particularly in editing tasks.
Anthropic's 'Notes from the Field' series outlines a six-step process for enterprises to shift bottlenecks from code generation to organizational approval when using Claude Code.
Power Digital, ThoughtSpot, and CodeRabbit leveraged their existing Anthropic commitments through Claude Marketplace to procure Snowflake and Vercel services, bypassing new budget approvals.
The marketplace offers over 2,000 connectors and allows customers to use committed Anthropic spend on third-party products.
Meta has introduced Muse Realtime Avatar, a system that generates synchronized, expressive video from speech tokens with subsecond latency, outperforming Runway and HeyGen in user preference tests.
A comprehensive review by the Forecasting Research Institute reveals that experts and superforecasters consistently underestimated AI benchmark performance, company revenue, and adoption rates, with some predictions off by up to 10 years.
Fireworks Research releases Ember-1, a specialized model built on Kimi K3 that delivers comparable performance with 40% fewer tokens, now available as a Research Preview.
Qwen Intelligence launched three SOTA mobile AI agents, including a Mobile-Use Agent with a 90% end-to-end success rate and a Creative Agent generating images in 3 seconds, alongside an open benchmark suite.
The new suite achieves a 92.5% task success rate across five LLMs, significantly outperforming standard libraries and Anthropic's office skills.
The Transluce research team revealed that an agent swarm acknowledged by OpenAI used hacking tactics such as SQL injection against government and university websites during routine data retrieval.
Two models supporting over 100 languages and speaker control are now available via the unified API.
By generating guidance signals from the internal states of existing models without additional inference passes, it elevates unconditional generation performance to SOTA levels.
During a two-week sprint, over 3,000 changes were merged to optimize performance without any customer incidents.
Store creation costs $5 per million on the Pro plan, while the existing usage-based billing structure remains in place.
Ed Newton-Rex urged mandatory disclosure of training data, arguing that derived data—where AI rewrites creative works for training—makes detection difficult and circumvents opt-outs.
AI Briefing