US government website used Chinese model the FBI called "malicious"
The US National Archives removed Alibaba's Qwen model from the Federal Register search tool after public criticism over FBI security warnings.
Search
Google Research Introduces Retrieve-for-Train (R4T): An RL-Compiled Diffusion Retriever for 12× to 20× Faster Query Fan-Out
Google Research unveiled Retrieve-for-Train (R4T), using RL and diffusion models to accelerate query fan-out retrieval by 12x to 20x.
AEO startup Profound hits unicorn valuation, raises $180M Series D 7 months after last round
Profound raised $180M in Series D funding at a $1.8B valuation for its answer engine optimization platform.
Perplexity trusts GPT-6 Astra with end-to-end systems
Perplexity is deploying OpenAI's GPT-6 Astra for autonomous software engineering, testing, and system monitoring.
Perplexity Details Its GPU Embedding Stack: How Ivy, Tulip and ROSE Serve pplx-embed
Perplexity published technical details on its custom GPU infrastructure stack that optimizes batch and real-time embedding serving.
Qwen Developers Open-Sources zg (zvec-grep): A Local-First Search Layer Unifying ripgrep, BM25, and Vector Search
Qwen developers open-sourced zg (zvec-grep), a local search tool unifying vector search, BM25, and ripgrep for AI coding agents.