Google Launches Open Knowledge Format, an AI Standard

Google has introduced Open Knowledge Format (OKF), a caller unfastened specification designed to standardize the measurement contented is packaged and shared pinch AI agents. Google Cloud Tech’s X station describes it arsenic a "vendor-neutral" standard, built to beryllium readable by some humans and machines without requiring caller devices aliases package to implement.

What is Open Knowledge Format?

OKF v0.1 uses a directory of markdown files pinch YAML frontmatter to found standardized conventions, making contented much compatible betwixt producers and AI agents. According to Google, a bundle of OKF documents is:

• “Just markdown:” each record is plain markdown that opens successful immoderate editor, renders connected GitHub, and tin beryllium indexed by modular hunt tools
• “Just files:” a bundle ships arsenic a tarball, lives successful immoderate git repo, and mounts connected immoderate filesystem
• “Just YAML frontmatter:” a short database of queryable fields covers type, title, description, resource, tags, and timestamp

Google has emphasized the simplicity of the format. There is nary compression scheme, caller runtime, aliases required SDK involved. 

For now, the reference implementations and tooling halfway connected Google Cloud, but the spec itself is unfastened and vendor-neutral, published connected GitHub nether the Apache 2.0 license. OKF v0.1 will germinate arsenic producers and consumers study what knowledge representations agents really need.

Why this matters for marketers

It's still early. OKF is an soul knowledge format for AI agents, not a hunt ranking aliases web-publishing signal, and it isn't tied to Google Search, YouTube, Maps, aliases different user products.

Google's statement is that OKF tin propulsion answers from different incompatible sources. Its ain examples are soul and technical: array definitions, metrics, and runbooks. For trading teams, the parallels are soul too, for illustration marque knowledge bases, reporting, income enablement, and training.

OKF isn't the only modular pointed astatine AI agents, and the differences are easy to confuse. An XML sitemap lists the pages connected your site. An llms.txt record points crawlers toward your astir useful content. OKF goes a measurement further and hands complete the knowledge itself, packaged arsenic files an supplier sounds directly. The quality that matters is direction: llms.txt and sitemaps look outward, toward the crawlers visiting your site, while OKF faces inward, toward an organization's ain agents.

For astir marketers, nary of this calls for action today. A published OKF bundle won't move your rankings this week aliases next. Ensure your contented is cleanable and well-structured, watch really the modular develops, and corroborate that AI crawlers tin already scope the contented you publish. If agent-facing standards drawback connected the measurement sitemaps and schema did, that foundational activity pays off.

How to hole pinch Semrush

Site Audit flags whether your tract is blocking AI crawlers and surfaces method issues that tin support your contented retired of AI-generated answers, including a missing llms.txt file. It runs connected each Semrush plan, truthful it's a low-lift spot to start.

To spot really that activity translates into visibility, the AI Visibility Toolkit tracks really often your marque gets cited crossed ChatGPT, Perplexity, Google AI Mode, and Gemini.

For endeavor teams, Semrush Enterprise AIO goes deeper, pinch AI visibility search crossed much LLMs and Crawler Profiles that simulate really bots for illustration ChatGPT and Googlebot entree and construe your site.