Brandkit Platform Tour: AI
AI in Brandkit
AI that augments (not replaces) humans and transforms your Brand Assets into AI readable formats.
We integrate with LLM's to power AI workflows (Auto-tagging, Auto-describing, Search Vector Generation, Semantic Search, Chatbots) in Brandkit. However AI augments and enhances human effort rather than replaces it. We use AI to powers workflows that automate busy-work (like tagging and describing your brand assets). It augments user added metadata —which in turn leads to better search results for all users. We even include an AI master switch to turn off AI altogether, if that's your preference.
And your data is never used for training by our LLM providers.
AI powered automatic tagging and describing of your images, video and posts
When you upload Images or Video, or add Posts, Brandkit, automatically tags and describes them with AI powered machine vision, using the API's of Large Language Models including OpenAI (aka ChatGPT) and Google Gemini.
You can add and edit a custom prompt for each Asset type in your Brandit account's Automation settings. You can then do things like ask for additional languages, or use humorous or formal language, or use localised slang words - all to match your brand's tone of voice.
Learn more here
(BETA)
AI powered semantic search
Once your Assets are classified, tagged, and described, we run AI again to embed your content metadata in a vector database, which is then used to power semantic (aka natural language) searches. Huh? Plain english please!. What this means for users is that they can search using vague'ish phrases or sentences and get sensible search results.
(Fig: Example of a Semantic Search for a "Romantic Dog")
Learn how Semantic Search works
Knowledge Base for AI
Brandkit automatically generates an AI friendly Knowledge Base that can be used as a dynamic (always up-to-date) knowlege source for AI chabots and AI agents.
That same knowledge is also exposed in a format ideal for LLM web crawlers to ensure your brand has the best opportunity to be discoverable and promptable in mainsteam foundation AI models like ChatGPT, Claude, CoPilot, Gemini and the like. And this is automatically Brand-Safe™ and secure by only exposing knowledge to web crawlers and bots that is publicy accessible.
(Fig: Example of an auto-generated LLMS.txt page)
Learn more
MCP
It's all well and good to have AI-powered workflows, chatbots, etc., in your Brandkit account - on your domain. But in the future (if not already), many people will be working in an LLM chatbot (e.g. ChatGPT, Claude, Gemini) and will never visit your Brandkit portal, or even your website.
The MCP protocol is the answer to "How do we connect our LLM chat to our own data and knowledge?"
MCP Server (Beta)
MCP server is how LLMs connect to a service like Brandkit. It's a new type of integration similar to an API but designed for AI.
Brandkit has an MCP server, and you can connect your LLM (e.g. Claude) to it. You can then ask your AI (e.g. Claude) to interrogate your Brandkit account, retrieve data, and take actions - based on the tools we build and expose via the MCP protocol.
MCP Apps (coming soon)
MCP Apps are a new standard for displaying UI components in LLM chats - so that when you ask your AI (e.g. Claude) to find images in your connected Brandkit, it can display the search results in a secure and sensible way directly in the chat experience.
Brandkit is currently testing and developing MCP Apps and will announce something soon.
FAQs: AI Features
What AI features does Brandkit offer?
We offer automatic tagging, content descriptions, and semantic search powered by AI.
Do I have to use AI features?
No, AI features are completely optional. You can disable them in your account settings.
What happens to my content when I use AI features?
Your selected assets are sent to OpenAI's API for processing. OpenAI generates tags/descriptions and returns them to Brandkit. OpenAI does not train AI models on your content.
How long does OpenAI keep my data?
OpenAI retains API data for up to 30 days for abuse monitoring, then automatically deletes it.
Is it secure?
Yes. Data is encrypted in transit and at rest. OpenAI is SOC 2 Type 2 and ISO 27001 certified.
Can I use AI features with sensitive or confidential content?
You should review your organization's policies on third-party AI processing before using AI features with sensitive content. Some organizations restrict this. You can disable AI features if needed.
Where is my data processed?
OpenAI's services are based in the United States. Your content may be processed there.
What if I don't want to use OpenAI?
You can disable AI features entirely. We're also monitoring alternative AI providers and may offer additional options in the future. Alternatively we may consider building a custom integration with an alternative AI provider for you.
Can I delete AI-generated tags and descriptions?
Yes, you have full control to edit or delete any AI-generated content in Brandkit.
Does using AI features cost extra?
Yes - but is cheap, especially when compared to time saved. See our Price page for detailed costs.