https://www.beri.net/article/anthropic-threat-report-eval-sandbox-litellm-prompt-injection-api-key-theft THE D[AI]LY BRIEF en 2026-09-12T20:01:42.484Z An Eval Sandbox Gave Up Its Keys. Your Gateway Holds Yours. https://www.beri.net/article/mcp-registry-random-sample-benchmark-duplication-server-start-rate THE D[AI]LY BRIEF en 2026-09-12T11:53:50.376Z 150 MCP Servers Never Started. No Benchmark Shows That. https://www.beri.net/article/papercut-agent-swarm-440-instances-12-domain-admin-lateral-containment THE D[AI]LY BRIEF en 2026-09-12T07:51:44.706Z 440 Print Servers Fell. Only 12 Reached Domain Admin. https://www.beri.net/article/model-router-buyers-guide-failover-not-judgment-2026 THE D[AI]LY BRIEF en 2026-09-12T03:48:20.892Z Model Router Buyer's Guide: Buy Failover, Not Judgment https://www.beri.net/article/analog-devices-alif-acquisition-ethos-u-vs-max78000-edge-ai-mcu-second-source THE D[AI]LY BRIEF en 2026-09-11T15:43:44.722Z ADI Is Buying Alif. Your Model Ports. Your Board Doesn't. https://www.beri.net/article/shopify-tailwind-labs-acquisition-ai-agents-open-source-funding-model THE D[AI]LY BRIEF en 2026-09-11T11:19:43.486Z Shopify Bought Tailwind. Your Agents Stopped Paying for It. https://www.beri.net/article/inference-cost-per-million-tokens-provider-pricing-compared-2026 THE D[AI]LY BRIEF en 2026-09-11T07:22:24.659Z Inference Cost per Million Tokens: Price the Task, Not the Rate