AI Briefing
GeekNews news
The latest AI and developer news from GeekNews, with short summaries.
Visit website →Alibaba Open-Sources 'open-code-review', an AI Code Review Tool Used Internally for 2 Years
Alibaba has open-sourced an AI code review tool that it has run internally for 2 years, catching millions of defects.
OpenAI Codex Record & Replay: Show a Task Once to Turn It into a Reusable Skill
A new feature has launched that records a user's workflow on Mac and turns it into a reusable AI skill.
South Korea's Telecom Giant SK Telecom at the Center of the Anthropic Mythos Controversy
SK Telecom's access to Anthropic's Claude Mythos has become controversial due to the U.S. government's export control measures.
Google Unveils Open Knowledge Format - A Knowledge-Sharing Standard for AI Agents
OKF, an open knowledge-sharing format designed to help AI agents efficiently consume fragmented internal knowledge, has been released.
OpenAI's 2025 loss nearly 8x higher, spending $34 billion
OpenAI's 2025 net loss surged 8x year-over-year, revealing the massive burden of R&D and infrastructure costs.
US puts entity-list blacklisting of over 100 companies including DeepSeek on hold over security risk classification
The US government has held off on adding over 100 Chinese AI and semiconductor companies, including DeepSeek, to its trade blacklist.
GLM-5.2 Takes the Top Spot Among Open-Weight Models on Artificial Analysis
Z ai's GLM-5.2 recorded the top performance score among open-weight models on Artificial Analysis metrics.
ktx - An Executable Context Layer for Data/Analytics Agents
ktx, a context layer that helps data agents accurately understand metrics and business knowledge in a data warehouse, has been released.
Is Meta Breaking Its Engineering Organization
Meta is forcibly restructuring its engineering organization into a data-labeling-focused ADO organization to secure AI competitiveness, triggering internal conflict.
Show GN: sfs - a shared brain filesystem for multiple AI agents
It's a filesystem tool that helps multiple AI agents collaborate by sharing context.
Agentic Code Review
As AI coding agents are adopted, code output has surged, but a bottleneck is emerging as review costs and defect rates rise together to verify that output.
Researchers say the Fable 5 controversy started not with a jailbreak but with 'fix this code'
Security experts claim and dispute that the access restrictions on Anthropic's model stemmed not from a jailbreak but from a simple code-fix request.
How Claude Code's Lead Designer Builds with AI [YouTube]
Introduces a real-world workflow that uses Claude Code to automate the product development pipeline, from idea exploration to writing PRs.
sandboxd - Open Source Sandbox Engine for AI App-Builder Products
An open source sandbox engine has been released that lets you build app-builder services powered by AI coding agents.
OpenRouter Fusion API
OpenRouter launched the Fusion API, which analyzes responses from multiple models in parallel and has a judge model synthesize them to boost performance.
[2026/06/08 ~ 14] AI/ML Papers Worth Checking Out This Week
A summary of 10 papers covering the latest AI research trends including autonomous agents, AI capability verification, and resource optimization.
Don't trust the big context window
This explains the gap between the advertised context window size of LLMs and the actual effective working range, along with efficient workflows.
Formal Methods and the Future of Programming
In the era of AI agent coding, the importance of formal methods is being highlighted as a tool to resolve the verification bottleneck.
Show GN: hera-agent-unity - An MCP Alternative for Controlling Unity via CLI (Zero Runtime Dependencies)
A lightweight tool that helps AI agents directly control the Unity editor and work based on actual measured data.
Rio de Janeiro's "in-house" LLM appears to be a merge of existing models
Allegations have surfaced that the Rio city government's supposedly self-developed model is actually the result of weight merging of existing models.