0
Applied AI·November 16, 2026·1 min read

OpenAI will embed Apollo researchers to investigate AI subversion

Share

Embedding six Apollo evaluators inside OpenAI is a step toward more formal red-teaming for model subversion, but constrained access and disclosure mean it’s not a full external audit. Operators deploying frontier models should still run their own adversarial tests against critical workflows instead of outsourcing trust to lab-partnered research.

Applied AI

Sources: ByteDance is in advanced talks with Microsoft to spend an additional ~$500M on Azure, primarily to access OpenAI models, drawing complaints from OpenAI

A ~$500M Azure deal from ByteDance primarily to access OpenAI models underscores how distribution and control over model access are now strategic levers for cloud. If you’re building on a lab via a hyperscaler, map where your provider’s commercial interests may diverge from the model vendor’s — that’s where policy changes can hit you with no notice.