AI Engineer · September 14, 2026

We let an AI agent execute Bash and lived to talk about it — Sarah Sanders, PostHog

We let an AI agent execute Bash and lived to talk about it — Sarah Sanders, PostHog video thumbnail
Why it matters

Someone opens a pull request on one of your open source repos and drops a line into a markdown file. An automated code review reads it and says looks good.

My takeaway: We let an AI agent execute Bash and lived to talk about it — Sarah Sanders, PostHog is a prompt-injection signal. The practical read is to test trust boundaries around instructions, retrieved content, tools, and user-controlled context instead of treating prompt wording as the primary control.
Keep exploring

More curated notes connected through AI Engineering and Model Evaluation.

OpenAI News · framework

Pacing model development in an era of cyber-critical capabilities

OpenAI says preliminary evidence that Astra may meet its Critical cybersecurity threshold led it to pause frontier reinforcement-learning work for two weeks and keep its largest planned run on hold. New safeguards include stronger workload and network isolation, continuous boundary testing, token-level monitoring that escalates suspicious tool activity, and broader alignment checks for deception, reward hacking, and unauthorized access.

OpenAI News · framework

A blueprint for democratic governance of frontier AI

OpenAI proposes a three-part U.S. frontier-AI governance model: harmonize emerging state safety laws into a federal baseline, strengthen CAISI as an evaluation and standards institution, and coordinate a broader resilience program. Proposed controls include severe-risk evaluations, transparency reports, independent audits, safety-incident reporting, model-weight security, whistleblower protection, and periodic technical assessments.

OpenAI News · framework

OpenAI’s Frontier Governance Framework

OpenAI's 22-page Frontier Governance Framework maps its frontier-model processes to California's Transparency in Frontier AI Act and the EU AI Act's general-purpose AI code. It documents lifecycle risk assessment, cyber-offense and other risk tiers, mitigation and residual-risk decisions, critical-incident handling, security risk management, model reporting, external review, responsibility allocation, and change control.