AIR Security emerged from stealth in early September 2026 with $50 million raised across two rounds in six months (Sequoia led a $10M, Greenoaks led a $40M). Its inline firewall screens the instructions, tools, plugins and MCP servers entering an AI agent's context for malicious prompts, excessive permissions and supply-chain risk; 20+ companies already use it.
AIR Security launches with $50M to firewall the AI agents enterprises deploy
If you ship AI agents or wire up MCP servers and third-party skills, treat the agent's tool supply chain as an attack surface and vet plugins/permissions before granting them access.
Source: TechCrunch
More that helps you.
Anthropic makes Claude Enterprise Admin API GA, adds agent site limits
Around Aug 19-22, 2026 Anthropic moved its Claude Enterprise Admin API (members, invites, groups and custom roles) out of beta, dropping the required beta headers, and let organiza…
IndiaAI Mission puts 38,000+ subsidised GPUs within reach of DPIIT startups
MeitY's official update (page last updated 16 September 2026) confirms that more than 38,000 GPUs for the common compute facility have been onboarded through the AI Compute Portal…
Anthropic adds auto permission mode to Claude Managed Agents
On September 10, 2026 Anthropic shipped an 'auto' permission policy for Claude Managed Agents that lets the server evaluate each agent or MCP tool call and run it, deny it, or paus…
Paris-based Arlequin AI raises EUR 28M for LLM-alternative topological neural networks
Arlequin AI closed a EUR 28 million Series A on September 10, 2026, co-led by redalpine and OTB Ventures with Bpifrance's Defence Innovation Fund and Xavier Niel joining. Its topol…
Get briefs like this tuned to you.
In the app, Founder Briefs are personalized to your country, industry and stage, and you can save the ones that matter.
See plans →