--- title: "One AI policy doesn’t fit your whole company. Now it doesn’t have to." url: https://daily.dev/posts/one-ai-policy-doesn-t-fit-your-whole-company-now-it-doesn-t-have-to--jtp69kkhn source_url: https://blog.kilo.ai/p/one-ai-policy-doesnt-fit-your-whole-org type: article source: "Kilo Blog" published: 2026-08-24T20:04:47.549Z updated: 2026-08-24T20:05:12.775Z tags: ["finops", "ai-governance"] reading_time: 3 upvotes: 0 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # One AI policy doesn’t fit your whole company. Now it doesn’t have to. **[Kilo Blog](https://daily.dev/sources/kilo-ai-blog)** · 3 min read · 0 upvotes · 0 comments ## Summary Kilo has introduced sub-org level governance for enterprise customers, letting a parent organization keep centralized budget and compliance control while each business unit sets its own AI model policies, providers, and inference routes. Admins can now spin up sub-orgs in one click, view all users across sub-orgs from a single screen, set per-sub-org model rules (e.g. restricting a European subsidiary to EU-only inference), and fund each sub-org separately with a flexible balance or Kilo Pass credits. Kilo's cost-saving features—Auto Efficient smart routing, open-weight model support, and Kilo Pass bonus credits—now apply at the sub-org level too. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Can I set different AI model policies for different business units within the same company using Kilo? Yes, Kilo now supports sub-org level governance where a parent account keeps centralized budget and compliance control while each sub-org sets its own approved models, providers, regions, and inference routes. For example, a European subsidiary can be locked to EU-only inference while other units keep the full model catalog on Kilo's gateway. _daily.dev helps engineering leaders track platform updates like this before rolling out org-wide AI policies._ ### How much cheaper is Kilo's Auto Efficient routing compared to using frontier models directly? Auto Efficient routes each coding task to the cheapest capable model, delivering 71% of frontier-level results at 72% lower cost. It still operates within whatever model pool a sub-org is restricted to, so tighter governance policies don't sacrifice the cost savings from smart routing. _teams comparing AI coding cost strategies can follow routing benchmarks like this on daily.dev._ ### What share of AI coding usage now goes to open-weight models instead of closed models? Open-weight models now handle 81% of real usage on Kilo, up from just 3% a year earlier. Mixing open and closed models lets teams get the most coding work done for the least spend, since Kilo's routing can choose across both categories per task. _developers weighing open versus closed model adoption can track shifts like this on daily.dev._ ## Similar posts on daily.dev - [More ways to control your AI coding spend](https://daily.dev/posts/more-ways-to-control-your-ai-coding-spend-wopjqthfb) · Kilo Blog · 0 upvotes · 0 comments - [Announcing KiloClaw for Organizations](https://daily.dev/posts/announcing-kiloclaw-for-organizations-kczmpgnnq) · Kilo Blog · 11 upvotes · 0 comments - [Kilo targets shadow AI agents with a managed enterprise platform](https://daily.dev/posts/kilo-targets-shadow-ai-agents-with-a-managed-enterprise-platform-coovhado4) · InfoWorld · 0 upvotes · 0 comments --- Tags: [#finops](https://daily.dev/tags/finops), [#ai-governance](https://daily.dev/tags/ai-governance) [View this post on daily.dev](https://daily.dev/posts/one-ai-policy-doesn-t-fit-your-whole-company-now-it-doesn-t-have-to--jtp69kkhn)