Collibra

10/27/2026

Your AI agents know a lot. That doesn't mean they understand your business.

They don't know what "customer" means at your company, which revenue metric is real or what data they're allowed to touch. A human would stop and ask, but an agent just guesses.

That guess is what we call the 'hallucination tax'. It shows up as wasted tokens, discarded outputs and endless manual reviews until leadership asks why the AI program is bleeding money.

At 5 agents, it's annoying. At 50, it's why the project gets canceled.

The fix requires governing the context layer across your entire stack, built directly into the foundation so agents get trusted data every time.

Read our new ebook, "The Missing Layer: Why context governance decides whether AI agents scale," for the framework to close that gap: https://ow.ly/iGXQ51018mo

10/26/2026

Among organizations with $100M+ in revenue, 96% say the root cause of failed AI initiatives almost always traces back to poor data foundations.

Scale was supposed to be the advantage. Instead, our latest research with The Harris Poll shows it's amplifying the problem most leaders assumed they'd already solved.

If you're leading AI at scale, you'll want to take a look at this. https://ow.ly/kr9G50ZZjtf

10/23/2026

London brought ambitious conversations and big ideas on governing AI at scale.

At the Collibra AI Summit, our CEO Felix Van de Maele welcomed data and AI leaders from across the UK and Europe to talk about what it really takes to trust AI in production, from clear visibility and accountability to governance that keeps pace with every agent.

Too often, ROI stalls because agents don’t know what your data means and no one can see what they’re doing. Collibra closes both gaps with governed context and control working as one system, so AI delivers value instead of rework.

Thank you, London. Watch the recap 👇

10/22/2026

Want trustworthy AI? Build the data foundation first.

Our co-founder Stijn asked Ulysses Demos, Head of Data Management and Protection Office at Red Sea Global, where teams should focus.

His advice is to build traceability early and take as long as you need, because everything else sits on top of it.

That way, when people questioned a new function or its return on investment, Ulysses could trace it back to a regulatory, legal or business requirement. Business requirements had to show ROI, which made those conversations much easier.

Hear how Red Sea Global moved past manual compliance to get ready for AI at enterprise scale: https://ow.ly/g94M50ZZ6gv

10/21/2026

Our CEO Felix Van De Maele recently asked a room full of data and AI leaders: "How many AI agents do you have in production? And what are they doing?"

Most couldn't answer either question.

That's where the industry is right now. Companies are encouraging teams to build and experiment, but no one is keeping count, and you can't govern what you don't know you have. Managing five agents is a staffing problem but managing a thousand is a systems problem, and you can't hire your way out of it.

Felix dug into this with Matt Marshall at VentureBeat's AI Week in New York. Better models won't solve it. Ask for your company's revenue and there's only one right answer, but agents miss it when they don't understand the business behind the data. Trusted AI at scale means knowing what you have, enforcing rules at runtime and keeping context accurate as the business changes.

We've spent two decades helping enterprises do this for data. Now we're doing it for agents.

Read more from Venturebeat: https://ow.ly/Htf150ZZtIQ

As AI agent fleets grow, enterprises need inventory, enforceable contracts and maintained business context to control risk, cost and reliability.

10/20/2026

Your agents are moving fast. Can you prove they’re safe?

We’re AIUC-1 certified, so we know a thing or two about agent safety, security and reliability. Our Maestro assistant stood up to thousands of real-world attack scenarios and completed an independent audit, making Collibra the first data and AI governance platform to hold the certification.

Now we want to show you what it takes to get there. On November 4, join Emil Bender Lassen from AIUC and our own Teresa Rojas and Wouter Mertens to learn how AIUC-1 is setting the bar for AI agents, and how the Collibra AI Command Center turns that standard into real control as your agent fleet scales.

📅 November 4 | 11 AM ET / 5 PM CET

👉 Register here: https://ow.ly/IN4P50ZXMoG

10/16/2026

Red Sea Global turned regulation, visitor trust and AI readiness into a business advantage. Ulysses Demos explains how his team built the data governance foundation behind it.

In our latest podcast episode, our Co-founder Stijn Christiaens chats with Ulysses about why governing the underlying data has to come first, and why AI agents need well-cataloged data from their own domain to become real experts.

Listen to the full conversation: https://ow.ly/7Kjb50ZXLE9

10/15/2026

9 in 10 organizations are deploying autonomous AI agents right now, but 76% of them have hit a wall trying to get from pilot into production.

According to a new survey conducted by The Harris Poll on behalf of Collibra, the real bottleneck has little to do with ambition or model quality. Most enterprises never built the context and control layer agents need to operate at scale.

Pilots can get by with messy data, but production can't. https://www.collibra.com/company/newsroom/press-releases/new-survey-from-collibra-by-the-harris-poll-finds-72-of-tech-decision-makers-feel-ai-initiatives?utm_source=linkedin&utm_medium=organic-social&utm_source=c90c570a-1869-45e8-a97b-d3f79b860319&utm_medium=organic_social&utm_campaign=

10/14/2026

Everyone at World Summit AI in Amsterdam was talking about speed.

Far fewer were talking about what happens when AI moves faster than the controls around it. Our Co-founder Stijn Christiaens was.

He addressed two of the most pressing issues facing enterprises today, the context and control gap and the resulting hallucination tax. AI cannot deliver full value until both are solved.

We brought the same conversation to the show floor. In our interactive workshop led by Wouter Mertens, attendees worked through real governance scenarios together and reached one conclusion. Governance is what lets you go faster with AI without crashing.

On stage, our customer Helvetia Insurance showed what that looks like in practice. By prioritizing the data governance foundation, the insurance group can scale what works and measure the business value.

Thank you to everyone who came by the booth to share what they're seeing and ask how we can help.

10/09/2026

An entire day devoted to getting AI into production, safely and at scale.

Thank you, New York. 🙌 This crowd of data and AI leaders brought countless questions and big ideas, and the featured speakers made every session count. Context, semantics, ontology and compliance dominated the conversation, all in service of AI that can be trusted in production.

The recap video is below 🎥 and if NYC was any indication, London on October 21 is going to be just as energizing. Register here: https://ow.ly/OUNC50ZW592