Home
Tools
Models
App
Prompts
News
Topics
Submit for inclusion
English
English 日本語 한국어 Português español Deutsch Русский Français 繁體中文 简体中文
option
Home Close
Home
Tools
Category Ranking Generate Cover Generate Image Skill
Models
Large Language Model Multimodal Model
App
Prompts
Prompts Image Prompts
News
News Flash News
Topics
Submit for inclusion
Submit App Submit Tool
Create an account Sign In
Change Language English Change Language
Settings
English EN 日本語 JA 한국어 KO Português PT español ES Deutsch DE Русский RU Français FR 繁體中文 ZH-TW 简体中文 ZH-CN
Home
Flash News
Content
JuanLopez
JuanLopez
April 2, 2026

Israeli security firm Adversa disclosed a critical flaw in Anthropic's AI tool Claude Code. A hard-coded limit of 50 sub-commands causes security rules to fail, switching from auto-reject to user prompts upon overflow. This allows attackers to embed long malicious command chains, potentially bypassing warnings especially in CI/CD environments. Experts urge immediate patching.

Israeli security firm Adversa disclosed a critical flaw in Anthropic's AI tool Claude Code. A hard-coded limit of 50 sub-commands causes security rules to fail, switching from auto-reject to user prompts upon overflow. This allows attackers to embed long malicious command chains, potentially bypassing warnings especially in CI/CD environments. Experts urge immediate patching.
0
0
Share Click to copy the link
Comments (0)
0/300
Submit
Author SamuelAllen SamuelAllen October 12, 2026
Chinese AI startup Moonshot (Kimi) is negotiating with former shareholders ahead of its Hong Kong IPO. Investors claim founders Yang Zhilin and Zhang Yutao misappropriated Recurrent.ai's assets, demanding $700 million in compensation. While Kimi denies the allegations, unresolved disputes may impact investor confidence in its governance and equity clarity. The company aims to settle these issues to mitigate compliance risks before listing, signaling its commitment to resolving internal conflicts prior to entering the capital market.
0
0
Author LawrenceRodriguez LawrenceRodriguez October 12, 2026
Tianjin publisher recalls book The Road to the End after AI editing prompt accidentally printed in text. The error occurred when software instructions were mistakenly included as main text during proofreading. The publisher apologizes, stating the author was not involved, and will destroy all copies for re-editing. This incident highlights risks of relying on AI in publishing without sufficient manual verification, serving as a warning for the industry regarding quality control and responsibility in the age of intelligent editing tools.
0
0
/live/7805
Author RalphRoberts RalphRoberts October 12, 2026
Tencent launched Penguin Micro, an AI creation matching platform within WeChat, connecting users needing digital content with skilled creators. Currently in early open phase during the National Day holiday, the service facilitates transactions for e-commerce visuals and short videos. Users post requirements, and the platform matches them with approved artisans. The workflow includes demand posting, matching, three-party communication for pricing and delivery, order placement, and settlement, all within WeChat. An AI assistant aids in clarifying needs to streamline communication. This initiative aims to standardize monetization for individual AI creators, though long-term success depends on mature pricing and delivery standards.
0
0
/live/7804
Author JuanLopez JuanLopez October 12, 2026
OpenAI increased GPT-6.1 Sol default context length from 272k to 500k tokens in Codex and Work environments. This 80% boost reduces compression and memory lapses for long tasks. Users can manually extend up to 1M tokens. Crucially, OpenAI removed extra charges for long context, lowering usage costs. This move aims to enhance user experience and compete in the large model market by making extended context a standard, cost-free feature rather than a premium add-on.
0
0
Author WillieAdams WillieAdams October 12, 2026
Amazon is in advanced talks to acquire AI startup Decart for approximately $7 billion. Decart specializes in cross-chip compatibility software that enables AI models to run seamlessly across different hardware manufacturers, addressing critical infrastructure flexibility. This potential acquisition, which represents a significant premium over Decart's recent $4 billion valuation, follows Anthropic's withdrawal from similar negotiations. The deal underscores Amazon's strategic push to lock in customers within its cloud ecosystem by solving hardware dependency issues and securing key computing power scheduling capabilities.
0
0
OR