InfoSecBulletin Cybersecurity for mankind
Integrating AI agents into corporate workflows can boost productivity, but it requires collaboration between AI developers and software firms implementing these agents. In response, OpenAI, Anthropic, Google, Microsoft, and Amazon have joined forces to establish the ‘Agentic Artificial Intelligence Foundation’ for AI agent standardization.
A report from The Information states that these companies will soon announce the foundation’s launch. Led by the Linux Foundation, the initiative aims to agree on ‘technical standards’ that enable diverse enterprise applications and AI agents to interconnect seamlessly.
Rapid7 Unveils 17 Zimbra Collaboration Suite Flaws Breaking Identity Boundaries
CISA launched two free tools Identify Security Misconfig in Microsoft 365 and Google Workspace
Apple iOS 27.2 Released Fixes 163 Flaws, Including Critical WebKit RCE
CISA adds PHP2Shell to KEV, PHP Releases New Security Updates to Fix Critical HTTP2Shell flaw
Cloudflare Unveils Emergency WAF Ruleset Update on Brotli flaw
Daily Cyber security update for 26.10. 2026
The Washington Post report
Ex-DOGE AI startup Cathedral Breaches Anthropic Internal Systems within 12 Hours
Daily Cyber security update for 25. 10. 2026
Daily Cyber security update for 24. 10. 2026
PHP shares new RCE flaw (CVE 2026-115560) with no patch but mitigation
AI agents used to function differently in various companies, leading to user complications. Each company and AI model had unique protocols, making task automation tricky. Moreover, varying specifications across firms posed security risks.
The foundation will first focus on standardizing ‘MCP (Model Context Protocol)’, a technology developed by Anthropic. MCP acts as a ‘connection rule’ that helps AI agents link with other apps—similar to how USB connects devices. With low adoption rates, the foundation aims to standardize AI agent functionalities. For instance, if a company connects ChatGPT to Slack via MCP, the AI could read employee conversations and answer managerial queries like “Who is responsible for a specific customer account?” Microsoft, Google, and Cursor have already integrated MCP into their products.
The foundation aims to standardize ‘Agents.md‘, a format for directing AI coding agents from OpenAI, and ‘Goose’, an AI agent by Block that works offline.
Analysts believe the partnership was prompted by the low profits from AI agents. Sales of AI agent products have struggled due to reliability and complexity issues. For instance, Microsoft has lowered its sales targets for AI products next year as market responses did not meet expectations.
