|

How Enterprises Can Leverage the ‘Linchpin’ of Internal Data

“AI is only as useful as the data you give it,” said Cohere’s Autumn Moulder, director of infrastructure and security

Photo of Cohere's Autumn Moulder
Photo via Cohere

Sign up for smart news, insights, and analysis on the biggest financial stories of the day.

Though enterprises are excited about the broad promises of AI, many are at a loss when it comes to how, exactly, they should leverage it. 

Along with figuring out where AI can provide the most value, many businesses are still navigating things like security, cost, control over data, and scaling use cases. The constantly evolving nature of this technology makes the task even harder for CIOs and IT teams.  

“I’ve seen a lot of leadership teams get excited about the broad potential with AI technology, but it’s crucial that teams focus on how their specific business can benefit,” said Autumn Moulder, director of infrastructure and security at Cohere. 

Moulder discussed with CIO Upside how AI fits into businesses, what people often get wrong about this tech and what’s next for the enterprise AI landscape. This interview has been edited and condensed for clarity.

What is something you think enterprises often miss in their AI strategies? 

The linchpin to seeing significant value is empowering your teams to both send real internal data and directly connect your systems to the AI models. It’s a step change in how your employees interact with computers. That’s where you’ll actually unlock the benefits of your AI strategy with productivity and efficiency gains. 

Many companies are rightly concerned about security and data privacy when they start considering how to directly connect their systems. It’s been such a common pain point in conversations with customers that, at Cohere, we ensure businesses can deploy our AI solutions privately in their own infrastructure or on premises for total control of their data. These challenges are heightened for highly regulated industries like finance, healthcare, critical infrastructure, or the public sector that deal with sensitive data. 

Another challenge centers around burdensome implementation of AI in organizations.  Ultimately, the build-it-yourself approach requires substantial resources, expertise, and maintenance costs. 

Many big AI companies focus on creating models for consumer products. What made Cohere realize that enterprise AI was the right move?

“We know AI will have a transformative impact on business and the way we work. I’m personally deeply motivated by helping people in their day-to-day work. While consumers have been experimenting with AI, enterprises have faced challenges adopting this technology for their workforce at scale. It’s a high-leverage problem to support enterprises in their adoption journey, and that’s why Cohere is focused on solving this challenge.

In order to accelerate wider adoption across industries, especially those traditionally slow to implement new technologies, trust is critical. Companies want technology that is reliable, easy to use, and secure. These are the core principles Cohere considers when designing our products. Our dedicated focus on the enterprise market enables Cohere to deliver AI solutions purpose-built for businesses that meet their customization, security, and privacy needs.”

What do you consider to be the biggest trends in enterprise AI for 2025? 

“As the technology continues to improve, we will start to see AI agents used at scale in enterprises. Core business functions like HR, finance, customer support, and IT will be equipped with customized agents to automate repetitive tasks and free up bandwidth to focus on the work that delivers the most value for their companies. 

Every time I engage with a partner, they ask how they can deploy agentic AI at scale. People see the vision; getting there is another question. Teams want help getting to the value in a low-risk way, and that’s why we engage in deep partnerships with businesses. We want our customers to get to production scale, and putting a real agent into production is much harder than just chatting with an AI model.

Let’s talk a bit about Cohere’s recent product, North. What inspired the decision to build a product focused on enterprise productivity? 

“Our goal with North is to help employees, regardless of their technical background, leverage AI to do their best work. With North, your finance or HR team can execute faster by creating custom AI agents to execute complex tasks like drafting documents or conducting research across your various internal data sources and applications. We achieve this by combining LLMs, search, and agents all into one platform that can seamlessly integrate with your existing workplace tools.

I’m very excited about North. I regularly work with our internal North system, asking quick questions that previously I would have pinged and interrupted my peers for an answer. When I need to discuss a decision with my teams, I’m better prepared. I find it’s improved my interactions overall with my colleagues across the business. It’s a mind shift and has changed the way I work. Getting every employee that kind of experience is what drove us to build North.”

North is also meant to meet strict privacy and security standards. Do you think enterprises that are adopting AI are taking security seriously enough? Why or why not? 

“The key to delivering value is only unlocked when you connect the system to all your internal data sources. Most businesses are very focused on security and privacy, and that’s been a huge barrier to seeing this workforce productivity improvement. That’s why we built North from the ground up to run in private – including air-gapped – environments. You can safely integrate all your sensitive data in one place. We see North as a way for businesses to control their agentic systems. It addresses an important gap in the market for highly secure and easy-to-implement AI solutions.”

In a recent LinkedIn post, Cohere CEO Aidan Gomez mentions the benefits of private deployment, like customization and security. Is this kind of AI adoption something that you anticipate to grow among enterprises? 

“Yes, we have seen a lot of interest from customers looking for private deployments in their own infrastructure or on premises. Companies need that level of control and security to get to production. When data never leaves your secured firewalls, it’s a lot easier to experiment with how you can use the technology and find the real value. This is especially true for regulated industries like finance, healthcare, manufacturing, and energy. 

We focus on offering customers maximum control and flexibility with various deployment options, including across all major cloud platforms. AI is only as useful as the data you give it, so Cohere’s focus on security and privacy enables organizations to build tools that are able to access their most valuable data sources.”

Cohere passed its five-year anniversary late last year. How has the AI landscape changed in those five years?

“The AI landscape is very competitive, and our models have continued to improve and expand their capabilities to deliver real-world business value. We have moved from the initial demo stage to securely deploying this technology at scale throughout large organizations like with our partners at Oracle across their Fusion and Netsuite applications. We are seeing real productivity and efficiency gains for customers across their core business functions. 

We continue to be focused on building on our models performance and capabilities with each generation and delivering a product fit to market needs. A big change we’ve heard from customers is the model alone is no longer enough. To make a big impact, we need fully end-to-end solutions like North that address the fundamental implementation issues with AI.”

How do you expect the AI landscape to change in the next year or so?

“I see AI agents helping workers offload routine tasks, opening up time to focus on what they enjoy most in their role. I see more companies focused on exactly where they can get specific value, and recognizing the need to secure and control these agentic systems. Cohere’s strategy of deep partnerships, industry-specific customization, and end-to-end AI solutions is focused on delivering tangible results for our customers in these areas in a shorter amount of time. With our unique focus on private environments, we’re excited to keep helping businesses unlock significant value.”