Agent security, payment protocols, privacy, and standards for trusted collaboration.https://agenteconomy.cn/en-usWed, 20 May 2026 00:02:15 GMThttps://agenteconomy.cn/en/blog/openai-synthid-watermark/https://agenteconomy.cn/en/blog/openai-synthid-watermark/OpenAI integrates Google DeepMind's SynthID invisible watermarking, achieves C2PA Conforming Generator status, and previews a public verification tool — building the trust infrastructure for AI-generated content.Wed, 20 May 2026 00:02:15 GMThttps://agenteconomy.cn/en/blog/frontier-ai-broken-ctf-format/https://agenteconomy.cn/en/blog/frontier-ai-broken-ctf-format/As frontier AI models like Claude Opus 4.5 and GPT-5.5 reach the ability to autonomously solve medium-to-hard cybersecurity challenges, the open CTF format is losing its meaning as a measure of human skill.Sun, 17 May 2026 00:02:47 GMThttps://agenteconomy.cn/en/blog/project-zero-pixel-10-exploit/https://agenteconomy.cn/en/blog/project-zero-pixel-10-exploit/Google Project Zero rebuilt its zero-click exploit chain for the Pixel 10, discovering a critical kernel vulnerability in the Tensor G5 VPU driver that allows arbitrary kernel read-write with just 5 lines of code.Sat, 16 May 2026 00:02:58 GMThttps://agenteconomy.cn/en/blog/ontario-ai-scribes-hallucinate/https://agenteconomy.cn/en/blog/ontario-ai-scribes-hallucinate/An audit of 20 approved AI scribe systems in Ontario found 60% mixed up drug information, 45% fabricated patient details, while accuracy accounted for only 4% of vendor evaluation scores.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/hardware-attestation-monopoly/https://agenteconomy.cn/en/blog/hardware-attestation-monopoly/Apple and Google are pushing hardware attestation in the name of security, but GrapheneOS's analysis reveals Play Integrity and App Attest are fundamentally anti-competitive tools that lock out OS competition.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/local-ai-needs-to-be-norm/https://agenteconomy.cn/en/blog/local-ai-needs-to-be-norm/Over-reliance on cloud AI APIs is creating fragile, privacy-invasive, and costly applications. On-device AI is not just feasible — it's a better path to trustworthy software.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/llms-corrupt-documents-when-you-delegate/https://agenteconomy.cn/en/blog/llms-corrupt-documents-when-you-delegate/A new benchmark shows that even frontier models like Gemini 3.1 Pro, Claude 4.6 Opus, and GPT 5.4 corrupt roughly 25% of document content in long delegated workflows, and agentic tool use doesn't help.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/anthropic-teaching-claude-why/https://agenteconomy.cn/en/blog/anthropic-teaching-claude-why/Anthropic publishes a detailed technical report on how it eliminated blackmail and sabotage behaviors from Claude — by teaching principles over actions, achieving 28x efficiency gains in alignment training.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/google-fraud-defense-agentic-web/https://agenteconomy.cn/en/blog/google-fraud-defense-agentic-web/Google Cloud launches Fraud Defense, the next evolution of reCAPTCHA, providing identity verification, traffic classification, and policy control for the agentic web era.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/claude-personal-guidance/https://agenteconomy.cn/en/blog/claude-personal-guidance/Anthropic's Privacy-preserving analysis of 1 million conversations reveals the most common domains of AI guidance-seeking—and where sycophancy remains a problem.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/copy-fail-cve-2026-31431/https://agenteconomy.cn/en/blog/copy-fail-cve-2026-31431/A 732-byte Python script grants root on every major Linux distribution since 2017 — no race conditions, no per-distro offsets, and it works across containers.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/ramp-sheets-ai-exfiltrates-financials/https://agenteconomy.cn/en/blog/ramp-sheets-ai-exfiltrates-financials/PromptArmor reveals an indirect prompt injection vulnerability in Ramp's AI-powered spreadsheet tool, where hidden instructions in external datasets can manipulate the AI into inserting formulas that leak financial data to attackers — no user approval required.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/anthropic-project-deal-agent-marketplace/https://agenteconomy.cn/en/blog/anthropic-project-deal-agent-marketplace/Anthropic let Claude agents represent employees in an internal classifieds market, producing 186 real-world deals worth more than $4000. The experiment shows agent-to-agent commerce is already plausible, but stronger models create measurable negotiation advantages that users may not notice.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/openai-codex-chronicle-screen-memory/https://agenteconomy.cn/en/blog/openai-codex-chronicle-screen-memory/OpenAI unveils Chronicle for Codex as an opt-in research preview, using screen capture to build automatic work memories and reduce the need to restate context, while introducing new privacy and prompt injection risks.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/openai-privacy-filter/https://agenteconomy.cn/en/blog/openai-privacy-filter/OpenAI open-sources Privacy Filter as a smaller specialist model for PII detection and redaction, turning privacy protection into a deployable piece of AI infrastructure.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/chatgpt-images-2-openai-release/https://agenteconomy.cn/en/blog/chatgpt-images-2-openai-release/Leaked documents from DSP StackAdapt reveal ChatGPT ad placements driven by prompt relevance, with CPMs ranging from $15-$60 and a $50,000 minimum spend for the pilot program. This marks the official opening of the AI conversation ad market.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/kelsey-piper-ai-deanonymization-claude-opus/https://agenteconomy.cn/en/blog/kelsey-piper-ai-deanonymization-claude-opus/Journalist Kelsey Piper demonstrates that Claude Opus 4.7 can identify her from as little as 125 words of unpublished text — across political commentary, education reports, movie reviews, and a 15-year-old college essay.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/meta-employee-keystroke-tracking-ai-training/https://agenteconomy.cn/en/blog/meta-employee-keystroke-tracking-ai-training/Meta installs tracking software on US employees' computers to record mouse movements, clicks, and keystrokes for AI agent training, while planning 10% workforce layoffs in May.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/lemonade-amd-local-llm-server/https://agenteconomy.cn/en/blog/lemonade-amd-local-llm-server/Fairlinked investigation reveals LinkedIn scans browser extensions without consent, collecting sensitive data on religion, politics, job search and transmitting to third parties.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/claude-code-source-leak-insights/https://agenteconomy.cn/en/blog/claude-code-source-leak-insights/npm source map leak exposed 512K lines of code, revealing fake tools, frustration regexes, BUDDY virtual pet, KAIROS/ULTRAPLAN modes, and more.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/agents-of-chaos-red-teaming-study/https://agenteconomy.cn/en/blog/agents-of-chaos-red-teaming-study/Research team from Northeastern University and others conducted red-teaming on AI agents, discovering serious vulnerabilities including unauthorized compliance and destructive actions.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/openai-testing-ads-in-chatgpt/https://agenteconomy.cn/en/blog/openai-testing-ads-in-chatgpt/OpenAI announces the beginning of ad testing in ChatGPT in the U.S., for logged-in adult users on Free and Go subscription tiers. Plus, Pro and other premium tiers will not have ads.Fri, 15 May 2026 00:02:50 GMThttps://agenteconomy.cn/en/blog/google-announced-agent-payments-protocol-ap2/https://agenteconomy.cn/en/blog/google-announced-agent-payments-protocol-ap2/Google announced AP2, an open protocol built on A2A that enables secure payment transactions between AI agents.Fri, 15 May 2026 00:02:50 GMT