Post-training
Everything done to a model after its initial pre-training on raw data: instruction tuning, reinforcement learning, fine-tuning on a company's own data, and safety training.
Pre-training teaches a model language and knowledge from a large corpus; post-training teaches it to follow instructions, use tools, reason step by step and behave within limits. For enterprises, post-training is where open-weight models are adapted to a specific domain, and it has become a product category in its own right: Palo Alto Networks' $4.5 billion purchase of Applied Compute in November 2026 is a bet that the post-training and inference layer is where control over enterprise AI will sit.
Get the brief in your inbox
One email every morning: the stories that mattered in AI, each with a one-line "why it matters" for your role. Free, no ads, unsubscribe any time.
Published daily by AI Briefing. We never share your address.