digitalNow 2026 Day  1  Recap: Who Holds the Keys to Your AI?

Day 1 of digitalNow 2026, Sidecar's annual conference on AI and associations, opened on Monday, October 26, at the Hilton Arlington Rosslyn at The Key, just outside Washington, D.C. Four main-stage keynotes came at AI from very different angles: data strategy, marketing, brand, and cybersecurity. Yet they kept circling back to one question. As AI becomes the most powerful tool associations have ever had, who stays in control?

Day 1 Key Takeaways

• Own your data, rent the intelligence. In the opening keynote, Blue Cypress founder Amith Nagarajan made the case for sovereign AI: unify your association's data in an environment you control and treat AI models as interchangeable.
• The old marketing playbook is broken. HubSpot CMO Kipp Bodnar shared Loop, the system he built and proved on HubSpot's own marketing engine, to replace it.
• Automation needs human moments. Raja Rajamannar, former Chief Marketing and Communications Officer of Mastercard, showed how AI agents and autonomous systems are reshaping brand engagement, and why purposeful moments of pause are growing in value.
• Think like a spy hunter. Former FBI operative Eric O'Neill explained how criminals are using generative AI, deepfakes, and the Dark Web, and how organizations can use counterintelligence thinking to protect themselves.

Opening Keynote: Amith Nagarajan on Sovereign AI

In his opening keynote, The Keys to the Kingdom, Amith Nagarajan, founder and chairman of Blue Cypress, argued that associations should never hand control of their data and AI to any single vendor. His framework, which he calls sovereign AI, rests on three ideas: own your data, rent the intelligence, and control the leverage.

Who Would You Trust With Your Association's Data?

Amith opened with a question instead of a statistic: who would you trust to safeguard your association's most important data for the long run? On screen were the leaders of the world's most prominent AI companies. His point was not about any of them as individuals. They are brilliant people building remarkable technology. But their companies have their own interests, and leadership, ownership, pricing, and terms all change over time. Your data and your members' trust have to outlast all of that.

The better person to trust, he said, is you. And he was clear that the same logic applies to every vendor, Blue Cypress included.

What AI Can Do for Associations Today

Before talking about control, Amith showed what is at stake. He introduced the International Cheese Federation, a fictional association built for the demo with 3,058 members across 641 organizations in 41 countries, plus events, courses, content, and years of history. That let him show everything working on realistic data without using any real association's records.

The audience followed one member, Mario Chen, through a unified member record, engagement signals, content activity, a renewal-risk prediction, and recommended next steps. Then they watched Mario talk with Sage, an AI agent that greeted him, knew his course history, and mapped his certification roadmap on the spot: three of five requirements complete, with an in-person cohort at the conference or a virtual cohort to finish. The takeaway was where it all lives: on the association's own data.

The Semi-Truck Problem

When a technology is this powerful, everyone will use it. The question is where it lives and who controls it. Amith described what is quietly happening across the sector: piece by piece, associations are backing up the semi truck of their data to a big tech company's loading dock and plugging it in. And the truck holds more than records. It carries:

• Member data
• Content and knowledge
• Context and memory
• Workflows and agents

The usual reassurance is "they don't train on our data." Under enterprise terms, that is often true. But training was never the main risk. The real risks are ownership, portability, and dependence: what happens when pricing, terms, or the company itself changes and everything you have built lives on someone else's dock? Anyone who has lived through a painful platform migration knows the cost of not owning your foundation. AI raises the stakes, because now it is not just your records on the line but the intelligence built on top of them. (For a deeper look at vendor incentives, see The Semi-Truck Problem: Evaluating the Economic Incentives of AI Vendors.)

The Three Principles of Sovereign AI

1. Own your data. Today, association data is scattered across the AMS, LMS, CRM, event platforms, and a growing pile of AI tools. The traditional fix is ETL: extract, transform into one standard schema, then load. Amith called the transform step the place where projects go to die. The alternative he described has no "T": copy each system as-is into an environment you control, let AI document every table and field, then let AI reason across all of it when you ask a question.
2. Rent the intelligence. Models keep getting better and cheaper, and the models riding just behind the frontier are extraordinary too, often faster and far less expensive. Treat models as interchangeable, use the best one for each job, and switch whenever something better or cheaper arrives. That is only possible if you own the foundation. (Related: The End of Model Monogamy.)
3. Control the leverage. You decide what data goes to which model, for what purpose, and with what permissions, and you keep the record. A controlled flow, not a firehose.

A Leadership Decision, Not an IT Purchase

Amith framed sovereignty as a strategy choice about the association's most valuable asset, one that belongs with the CEO, the leadership team, and the board. He drew a line between running the business and building something new. Internal IT teams keep the lights on, and that is exactly what they should do. IT is not failing; IT is running. But build work never fits in the margins of run work. Associations need both, and they need to decide what stops: ad hoc uploads into whatever tool is handy, critical workflows built inside a single vendor's platform, and pilots that do not build on a foundation they own.

His advice on timing: the window is open now. Every month spent building on someone else's dock makes the move home harder. Pick one high-value use case, build it on your own foundation, and expand from there.

Four Questions to Ask Any AI Partner

1. Where does your data live, and who decides?
2. Can you switch models freely?
3. Who owns what you build?
4. Can you take it all and leave?

Amith encouraged attendees to ask these of any partner or platform, including Blue Cypress. As one example of how a platform can answer them, he pointed to MemberJunction, an open-source, model-agnostic data and AI platform that is free for associations. Associations bring their own model keys, can run it in their own cloud or a partner's, and can take their data and everything they have built with them if they ever leave.

He was also candid that he is biased, and that he cares more about associations getting this right than about which partner they choose. A different partner, your own team, or some mix can all work. The only wrong choice is the one nobody actually makes: drifting into dependence one convenient tool at a time.

He closed where he started. The leaders of the big AI labs are building the most powerful technology any of us will see in our lifetimes, and associations should put it to work for their members. But many associations have been around for decades, some for over a century, and they need to be around for decades more. The job of holding the keys cannot be outsourced, not to the AI labs, not to Blue Cypress, not to anyone. Associations already hold the keys. The call to action was to use them.

Kipp Bodnar: AI Broke the Old Marketing Playbook

HubSpot Chief Marketing Officer Kipp Bodnar told the room that AI broke the playbook every marketer was running, and that until now nothing has replaced it. His answer is Loop, a new system he built and proved on HubSpot's own marketing engine. Kipp's core argument: the marketers who learn and adapt fastest will win, do better work, and actually enjoy the job again.

Loop is also the subject of Kipp's new book with Kieran Flanagan, Loop: Outlearn. Outmarket. Outgrow. For association marketing and membership teams, the message lands hard. As members and prospects increasingly turn to AI assistants instead of search bars, the way associations get found and stay relevant is changing, and the teams that build fast learning cycles into their marketing will adapt first.

Raja Rajamannar: Marketing Journey Into the Future

Raja Rajamannar spent 12 years as Chief Marketing and Communications Officer of Mastercard, where he evolved the Priceless campaign into a multisensory brand experience. He is now a senior fellow at Mastercard and the author of the Wall Street Journal bestseller Quantum Marketing.

His session explored how marketing must evolve as exponential technologies, from AI agents to autonomous systems, reshape how people make decisions and engage with brands. He highlighted a rising insight in consumer behavior: the growing value of purposeful moments of pause that make interactions feel more tangible and real. His roadmap for an intelligent, interconnected future is to balance advanced automation with human-centered design so experiences stay memorable and emotionally resonant.

For associations, that is a useful reminder. The conferences, communities, and peer connections associations already run are exactly the kind of human moments that become more valuable as more of daily life gets automated.

Eric O'Neill: Spies, Lies, and Cybercrime

Eric O'Neill began his career as an undercover operative in the FBI's counterintelligence trenches, where his work brought down Robert Hanssen, one of the most damaging spies in U.S. history. That story became the film Breach, and Eric has since written two bestselling books, Gray Day and Spies, Lies, and Cybercrime.

His session took attendees into the world where foreign spies, deepfakes, and Dark Web criminals use generative AI to steal from organizations, with cybertheft on track, by his estimate, to take more than $20 trillion from organizations like the ones in the room. He covered:

• Why cybercrime is the fastest-growing business on earth
• The connection between espionage and Dark Web cybercrime
• Counterintelligence techniques to protect yourself and your organization
• How to think like a spy hunter

Paired with the morning's keynote, the lesson was clear. As associations bring their data together to put AI to work, protecting that data becomes just as much a leadership responsibility as using it.

Closing Keynote Panel and Q&A

Day 1 wrapped with Amith Nagarajan and Eric O'Neill together on the main stage for a keynote panel and audience Q&A, bringing the day's two biggest themes, data sovereignty and security, into one conversation.

Coming Up on Day 2

Tuesday, October 27, continues on the main stage:

• 8:45 AM ET: Dave Dellin, CEO of the American Quarter Horse Association, on AI From a Non-profit CEO's Perspective, including how the American Paint Horse Association's Becca Bot cut routine phone calls by 28%
• 10:05 AM ET: Career diplomat Danny Stoian on Punching Above Your Weight Class
• 1:00 PM ET: Safi Bahcall, author of Loonshots, on Five Rules for Innovating in Turbulent Times
• 3:15 PM ET: Keynote panel and Q&A

On Wednesday, Sidecar's Mallory Mejias closes the keynote lineup with Auditioning for Relevance: Winning a Recurring Role in Your Members' Lives.

Frequently Asked Questions

What is digitalNow?

digitalNow is Sidecar's annual conference at the intersection of AI and associations, built for association executives and their teams. digitalNow 2026 runs October 25 to 28 at the Hilton Arlington Rosslyn at The Key in the Washington, D.C. area.

What is sovereign AI for associations?

Sovereign AI is the approach Amith Nagarajan laid out in the digitalNow 2026 opening keynote: the association keeps ownership and control of its data, the intelligence built on it, and the choice of which AI models to use, rather than moving those assets into a single vendor's platform.

What are the three principles of sovereign AI?

Own your data by unifying it in an environment you control. Rent the intelligence by treating AI models as interchangeable and switching to better or cheaper ones as they arrive. Control the leverage by deciding what data goes to which model, for what purpose, and with what permissions.

What questions should an association ask an AI vendor?

Where does our data live, and who decides? Can we switch models freely? Who owns what we build? Can we take it all and leave? If a partner answers all four the right way, the association stays in control no matter which partner it picks.

Who spoke on Day 1 of digitalNow 2026?

Day 1 keynotes featured Amith Nagarajan (Blue Cypress), Kipp Bodnar (HubSpot), Raja Rajamannar (Mastercard), and Eric O'Neill (The Georgetown Group), followed by a keynote panel and Q&A with Amith and Eric.

Explore the full digitalNow 2026 agenda and speakers