Box empowers developers with new AI model flexibility and AI developer platform integrations
As AI continues to rapidly evolve with new and more powerful and efficient models coming out practically every day, the challenges and opportunities presented to developers in this space also continues to grow.
Developers are looking to build a wide range of applications and solutions with enterprise-grade AI, aiming to enhance efficiency, productivity, and decision-making within organizations. At Box, we're integrating cutting-edge LLM capabilities directly into your Box experience. We're focusing on understanding specific LLM strengths and enabling customer choice to empower your content journey, instead of simply using a one-size-fits-all approach.
This is why we are excited to announce choose-your-own-model (CYOM) in Box AI APIs, enabling the flexibility for developers to now choose their model from a best-in-class selection for their API use cases, including models hosted by Google Cloud and Microsoft Azure.
By leveraging CYOM via API, developers can:
- Choose their models: Box AI API now provides developers with the ability to select the models that best align with their specific use cases and requirements, ensuring high quality and consistent experiences tailored to each developer’s needs.
- Integrate with applications: Box AI API is designed to integrate seamlessly with new and existing experiences and workflows. Developers can easily incorporate AI capabilities, including question and answer, content generation, and metadata extraction into their applications, enhancing functionality and improving user experiences.
Box will support developers to choose from any of the listed supported models when leveraging the Box AI Platform APIs, including:
- Google's Gemini Flash on Vertex
- Azure OpenAI GPT4.0
- Azure OpenAI GPT3.5
“Enterprises use AI to build powerful experiences that extract insights from unstructured data and content stored on Box,” said Dr. Ali Arsanjani, Director, Applied AI Engineering at Google Cloud, “By enabling developers to choose their preferred models, customers can build more valuable and reliable experiences, while leveraging the millions of files in their organization.”
In addition, as Box continues to enable developers to build AI applications that leverage their content to accelerate business processes, we are excited to announce new integrations with AI developer platforms.
- Google Vertex AI Search: Use the Box connector for Vertex AI Search to create agents and apps that are grounded in your enterprise content on Box.
- LangChain: Use the Box integration package in LangChain to integrate the Box document loader and retriever into agents and applications.
- LlamaIndex: Combine your Box content and LlamaIndex to build Enterprise-grade AI tools with LlamaCloud and the Box connector, build powerful and intuitive AI agents with the Box agent tools or index your files with the Box reader.
- Microsoft Copilot Studio: Use the Box connector for Microsoft Graph and Microsoft Power Platform to create and customize agents in Microsoft Copilot Studio with your enterprise content on Box.
- Pinecone (coming soon): Use the Box connector to bring Box data into Pinecone vector databases to enable low-latency search in your applications and agents.
"Developers are using AI to create smarter workflows and content-aware apps. Integrating enterprise data is key to making these tools useful to organizations," said Harrison Chase, CEO of LangChain. "The new Box integrations allow businesses to immediately incorporate their content into LangGraph agents and LangChain apps. We're excited to see what developers build on top of their Box data."
As with all AI offerings, Box AI continues to stay committed to providing responsible and secure use of AI on enterprise content, and the same goes for CYOM in the Box AI Platform APIs. Box AI incorporates robust security measures, including:
- Data Privacy: Box AI is committed to protecting user data privacy. We adhere to strict data privacy regulations and implement measures to safeguard sensitive information.
- Access Controls: Box AI leverages secure RAG which ensures that only authorized users have access to the right content and usage of AI.
- Compliance: Box AI is designed to comply with industry standards and regulations, such as GDPR and HIPAA. This provides organizations with the confidence that their AI deployments are compliant with relevant laws and regulations.
The field of AI is constantly evolving, with new technologies and techniques emerging at a rapid pace. Box AI is committed to staying at the forefront of these advancements. We continuously invest in research and development to ensure that our platform remains up to date with the latest AI trends. This commitment to innovation means that developers can rely on Box AI to provide them with the tools and capabilities they need to stay ahead of the curve.