Context by Cohere
Introducing Coral, the Knowledge Assistant for Enterprises

Introducing Coral, the Knowledge Assistant for Enterprises


Today, we're excited to introduce Coral, a knowledge assistant for enterprises looking to improve the productivity of their most strategic teams.

As work environments have evolved, so have the ways workers seek out information and complete their jobs. Employees want to ask questions in natural language and receive relevant, helpful answers – just like they can do with their colleagues.

But with the growing volume of information in today's enterprises, it's becoming increasingly difficult for workers to find the answers they need to stay productive. According to a McKinsey report, employees can spend up to 20% of their day searching for information, impacting the speed and performance in which they do their jobs.

This is where Coral comes in. Coral is an enterprise chatbot that can converse with users to help them complete their business tasks. Coral is powered by Cohere's Command model that is trained with chat, reasoning, and writing abilities. It's then customized for companies by augmenting its knowledge base with data connections and deployed privately to protect sensitive data.

Redefining Productivity

With existing consumer chatbots, we've seen early evidence of improved productivity. These research studies show that the time it takes for brainstorming and drafting communications is improved by up to 50%.

For business, we see extensive demand for knowledge assistants to apply across all business functions. While industries, domains, and jobs may differ, all teams still want to get their jobs done faster:

  • Knowledge workers such as account executives, analysts, consultants, engineers, lawyers, etc., have defined roles they need to perform efficiently and at high quality. While specific tasks may vary across different teams, all must research, analyze, and recommend as part of their core jobs. With the promise of intelligent chatbots, knowledge workers already see the value of having an easy technology interface to support their day-to-day functions.
  • Customer support departments need product information quickly and accurately. Empowering customer support agents with an internal chatbot that knows product and support details can resolve cases faster. An NBER study found that customer support agents having access to such an assistant helps resolve 14% more cases on average. Other benefits included improved customer sentiment, reduced requests for managerial intervention, and improved employee retention.

Enterprises Require Data Privacy, Want Trust

Despite the overwhelming excitement for chatbots in the workplace, data privacy and placing trust in generative AI are substantial barriers to adoption. Notably, many companies have banned consumer chatbots inside their businesses because of the risk that sensitive data will be leaked outside the company. Existing consumer chatbots often require data to be sent to an external managed service for it to work. It's a considerable risk for businesses that deal with sensitive internal or customer data.

Also, hallucinations are a part of utilizing generative AI that are difficult to identify. For better or for worse, LLMs can sound confident in their responses, and a confident hallucination can escalate to spectacular misunderstandings. Business users want the ability to verify the responses of these chatbots.

The Knowledge Assistant for Enterprises

Cohere is developing an enterprise-grade chatbot for business use. We're working with market leaders to design and implement these knowledge assistants tailored to their most strategic teams. Coral is:

Conversational: With chat as its primary interface, Coral understands the intent behind conversations, remembers the history, and is simple to use. Knowledge workers now have a capable assistant supporting their business tasks, one that can research, draft, summarize, and more. For example, a financial analyst can ask for an overview of a new market, identify the major players, and generate a financial overview all within the same conversation!

Customized: Out-of-the-box chatbots don't know your business. Your industry, domain, and company nuances are essential when completing business tasks. Customers can augment Coral's knowledge base through data connections. Coral has 100+ integrations ready to connect to data sources important to your business across CRMs, collaboration tools, databases, search engines, support systems, and more.

Grounded: To help verify generations, Coral can produce responses with citations from relevant data sources. Behind the scenes, our models are trained to seek relevant data based on a user's need, even from multiple sources. This grounding mechanism is essential in a world where workers need to understand where information is coming from in a consumable way.

Private: Companies that want to take advantage of business-grade chatbots must have them deployed in a private environment and aligned with the principles that make technology safe and data secure. The data used for prompting and the chatbot's outputs will not leave a company's data perimeter. Cohere is cloud-agnostic and will support deployment on any cloud.

What Companies Are Saying

“Cohere is helping organizations accelerate their AI initiatives with cutting-edge knowledge augmentation capabilities,” said Greg Pavlik, SVP Oracle Cloud Infrastructure. "With our partnership with Cohere, Oracle can provide native generative AI-based features to help organizations use their own data to advance essential business functions, improve decision-making, and enhance customer experiences.”

"The combination of LivePerson's industry-leading conversational platform and AI with Cohere's Coral will help deliver custom LLMs for customer engagement built on the enterprise's specific needs, goals, policies, and data," says Joe Bradley, Chief Scientist from LivePerson. "Coral's knowledge augmentation capabilities will connect our solutions to additional data sources to keep LLM-powered conversations grounded, factual, and generate outputs that match enterprise needs in real-live use cases."

“We're thrilled for enterprises to be able to seamlessly and securely pair Elasticsearch with Cohere's Coral to achieve better business outcomes," said Matt Riley, General Manager for Search at Elastic. “Elasticsearch AI, coupled with Coral, enables organizations to take advantage of all their structured and unstructured data to increase employee efficiency."

Coral is currently in private access with a select group of customers. If you're interested in participating in the program, talk to our team at

This announcement was drafted using Coral, with additional refinement by humans.

Keep reading