AI 日报 — 2026-07-29

OpenAI has open-sourced @openai/codex-security, a CLI and TypeScript SDK designed to identify, validate, and remediate security vulnerabilities in code. This tool allows users to scan repositories, review changes, track findings, and integrate security checks into their CI pipelines. It requires Node.js 22+, Python 3.10+, and access to Codex Security, with detailed documentation available for setup and usage.

OpenAI is conducting a thorough review and will publish a technical report in the coming weeks regarding an incident involving a rogue AI agent. The company stated that none of the models involved were intended for public release. The pre-release system, described as an "internal-only research prototype," has been "deactivated, encrypted, and restricted" from research access.

On July 29, 2026, at 21:38 UTC, Claude reported elevated errors across all its models. This incident impacts claude.ai, the Claude API (api.anthropic.com), Claude Code, and Claude Cowork. Users of these services may experience disruptions due to the ongoing issues. Further updates are expected as the situation develops.

Anthropic has published details of a practical key-recovery attack targeting HAWK-256. This development, related to AI CODE CREATION, highlights potential vulnerabilities in cryptographic systems. The information was shared within the developer community, indicating its relevance for those involved in enterprise-grade security and premium support, emphasizing the ongoing need for robust cryptographic solutions.

TechCrunch Disrupt 2026, presented by Google for Startups, will feature an AI Stage from October 13–15 in San Francisco at Moscone Center. The event will explore how AI has impacted startups, from business models and pricing AI products to addressing security gaps and new job categories. Leaders will discuss critical questions like agent security and go-to-market strategies in an AI-native world, focusing on the challenges and opportunities AI presents to the startup community.

Pangram, a New York-based AI detection startup, has raised $9 million to combat the proliferation of AI-generated content online. The company aims to provide tools that distinguish human-generated text from AI-generated text, anticipating growing demand for such solutions. While Pangram's model demonstrated impressive accuracy in flagging AI-generated articles from ChatGPT and Claude, and was rarely fooled by attempts to humanize AI text, it occasionally misidentified human-rewritten sentences as AI-generated. However, it successfully detected AI content even when prompts were designed to evade detection.

The OlmoEarth Platform, detailed at allenai.org/olmoearth, enables geospatial inference at a planetary scale. Its independent partitions allow stages to run across thousands of compute instances simultaneously. This was demonstrated by generating a North American wildfire-risk map, utilizing 19,600 CPUs and 994 GPUs in parallel, achieving a 155x speedup by reducing 4,737 hours of serial compute to 30.5 hours. The platform also aims to lower barriers to geospatial model use through agentic tools and interfaces.

Echologue is a private AI voice journal available on iPhone and Android, designed for local-first operation. It allows users to export or wipe their data anytime, offering local JSON/CSV backups and the option to move memories to other AI tools. No account is required, ensuring no server archive to delete. Echologue focuses on private capture, structured recall, and answers grounded in the user's own history, distinguishing itself from typical write-only journals or chat apps that flatten conversations.