Cohere has introduced North, an enterprise-focused AI platform designed to address the challenges of adopting AI in regulated industries such as finance and healthcare.

By prioritizing privacy, security, and customizability, North positions itself as a robust alternative to Microsoft 365 Copilot and Google Vertex AI. The launch show the increasing focus on enterprise AI solutions that offer both advanced capabilities and the ability to meet stringent data protection requirements.

What is Cohere North?

Cohere North is an AI workspace designed to streamline enterprise workflows while maintaining rigorous security standards. It combines generative AI capabilities, advanced search tools, and customizable AI agents, providing businesses with an end-to-end solution for integrating AI into their operations.

A standout feature of North is its Compass Search System, a multimodal tool capable of processing and retrieving information from diverse data formats, such as documents, spreadsheets, and images.

By utilizing Cohere’s proprietary Embed and Rerank models, Compass ensures that search results are both relevant and precise. According to internal benchmarks, Compass reduces task completion times by over 80%, significantly improving productivity.

Another key feature is North’s ability to create and deploy customizable AI agents. These agents can automate routine tasks across various functions, such as HR, finance, and IT, without requiring programming expertise.

For instance, HR professionals can use North to develop agents to automate onboarding processes, while finance teams can create agents to streamline data analysis.

North also supports private cloud deployments and air-gapped systems, ensuring that sensitive enterprise data remains secure and compliant with regulatory standards. This feature makes North particularly appealing to industries like healthcare and finance, where data privacy is paramount.

At its core, North is powered by Cohere’s Command R large language model, which is optimized for real-world generative tasks. The model balances efficiency and accuracy, enabling users to generate reports, summarize meetings, and manage other complex workflows seamlessly.

How North Compares to Microsoft 365 Copilot

Microsoft 365 Copilot is deeply integrated into the Microsoft ecosystem, offering productivity enhancements within tools like Word, Excel, and Teams.

Its features include Business Chat, which allows users to manage tasks like meeting scheduling and email drafting across applications, and the Prompt Gallery, a repository of pre-built and customizable prompts.

However, Microsoft Copilot’s reliance on its ecosystem limits its flexibility. It is most effective for businesses already heavily invested in Microsoft 365 and struggles to adapt to workflows outside of its applications.

Additionally, Copilot requires an active internet connection, which can pose challenges for organizations operating in restricted environments.

While Copilot excels in enhancing productivity within the Microsoft suite, it lacks the focus on privacy and secure deployment that defines Cohere North. This distinction could make North a stronger choice for regulated industries requiring robust data protection.

However, while North offers flexibility to integrate with various tools, it lacks the pre-built, deep integration of Microsoft 365 apps. Businesses already entrenched in the Microsoft ecosystem may find North less convenient or require additional effort to connect it to existing workflows.

Google Vertex AI: A Versatile but Complex Alternative

Google Vertex AI is a unified platform designed for building and managing machine learning models. It offers tools for custom model training, MLOps tools (for monitoring and scaling), and access to advanced pre-trained models like Google’s Gemini.

Vertex AI’s strength lies in its versatility; it caters to a wide range of users, from AI beginners to experienced developers. Features like Model Garden (access to pre-trained models like Gemini) and AutoML (simplified model training) make it suitable for a wide range of AI applications.

Despite its comprehensive toolset, Vertex AI’s complexity can make it challenging for businesses seeking a straightforward solution. Additionally, its lack of emphasis on privacy and secure deployment makes it less suitable for industries with stringent regulatory requirements.

Cohere North instead prioritizes secure deployments and user-friendly customization, making it a better fit for businesses that value privacy and operational simplicity. However, North so far lacks the depth and breadth of tools Vertex AI provides for developers and data scientists.

Cohere North’s Unique Selling Points

Cohere North’s combination of privacy-focused deployment, customizable workflows, and advanced AI capabilities distinguishes it from competitors. Its ability to operate in air-gapped environments ensures maximum data security, a feature that neither Microsoft Copilot nor Google Vertex AI offers.

Cohere says that North outperforms both Microsoft’s Copilot and Google Vertex AI in terms of retrieval-augmented generation (RAG) accuracy “across Finance, HR, Customer Support and IT benchmarks” when measured with the Llama Index (LI).

LI is a common industry metric that assigns a score to a completion based on its comparison with a ground truth answer. RAG combines LLMs with external knowledge bases to improve their outputs.

Another critical advantage is North’s early adoption by prominent organizations. The Royal Bank of Canada (RBC) has partnered with Cohere to develop a tailored version of North for the financial sector, highlighting the platform’s potential for industry-specific applications.

North is currently available through an early access program, with a focuson sectors such as healthcare, finance, and manufacturing. Its partnership with RBC signals its potential for broader adoption, particularly in industries where data protection and compliance are non-negotiable.