The interview

Rob, let’s start at the top. When you look at where family offices are today versus where they will be in 2035, what changes most?

Robert E. Mallernee, Founder & CEO, Eton Solutions: The operating model changes fundamentally. Today, most family offices run on a mix of specialised software and human labour that stitches the systems together. In 2035, the offices that have kept pace will be running on AI-native platforms where the routine mechanical work is handled by autonomous agents and the professionals in the office spend their days on analysis, judgment, and principal-facing work. Both offices will look like family offices from the outside. Inside, they will operate completely differently.

AtlasFive®, the platform we have been building at Eton for over a decade, was designed for this operating model from the ground up. Not retro-fitted around AI. Purpose-built for it. That distinction is going to matter more every year.

You use the phrase “AI-native” a lot. What do you mean by it specifically?

Rob Mallernee: I mean a platform that was architected assuming AI would do most of the routine operational work, rather than a platform that was designed for human operators and had AI added on top. Those are two completely different design starting points, and they produce very different results in production.

An AI-native platform has an integrated data model that AI can reason across. It has a knowledge graph that captures relationships between entities, not just documents. It has autonomous agents that can execute end-to-end workflows under human oversight, not just answer questions. And it has governance woven throughout, not bolted on as a disclosure. AtlasFive® has all four. Most platforms in our category have some of them and are working on the rest.

What is the single most important thing family offices should be evaluating when they choose a platform in 2026 or 2027?

Rob Mallernee: Architecture. Not features. Features change every quarter as AI capabilities evolve. Architecture takes years to build and defines what the platform will be capable of in 2029 or 2030. Family offices making a platform decision this year are choosing more than a tool. They are choosing which architectural bet to make for the next decade.

Ask a few specific questions. Does the platform hold an authoritative record, or does it read from systems held elsewhere? Does the AI understand entities and business rules, or primarily text? Can the AI execute within the system of record, or only produce answers? Is the governance certified against an international standard, or asserted in marketing materials? Those four questions separate the platforms that will still be relevant in 2030 from the ones that will not.

Let’s talk about trust. You have said publicly that trust is the currency of the AI era in wealth management. What do you mean?

Rob Mallernee: Family offices handle some of the most sensitive financial and personal data in the world. As AI systems take on more of the operational work in that context, the question of whether the family can trust the AI becomes a direct fiduciary concern. Not “does the AI work” as a technical question. “Do we trust it enough to let it act on our behalf” as a governance question.

Trust in the AI era is earned through three things. Governance that is independently audited rather than self-attested. Data sovereignty that gives the family cryptographic control of its own data. And transparency about what the AI does, why, and under what oversight. We invested heavily in all three at Eton because we saw this becoming the centre of the buying conversation years before it did. ISO 42001 certification is the concrete result of that investment. So is our Bring Your Own Key encryption. Neither is a marketing feature. Both are load-bearing to how a family office should think about placing operational trust in a technology partner.

ISO 42001 keeps coming up in your conversations with family offices. Why does it matter more than the certifications the industry has had for years?

Rob Mallernee: Because it is the first international standard specifically designed for how AI is governed, and it is now the difference between a platform that has a defensible answer to “how do you govern your AI” and a platform that does not. SOC 2 Type II and ISO 27001 tell you an organisation handles security and information governance well. They do not, on their own, tell you anything specific about how the AI is developed, monitored, and controlled.

ISO 42001 fills exactly that gap. It certifies that the organisation has documented AI governance, risk assessment for each AI system, data governance for AI training, human oversight controls, transparency and explainability, incident management, and continual improvement. That is the full stack of what a family office board should want to know about the AI operating on its behalf. Eton is one of the first family office platforms globally to hold the certification. I expect the rest of our category to be pursuing it within the next 18 months, because family office boards are going to start asking for it in RFPs.

A lot of family offices are worried that AI will replace their people. What do you tell them?

Rob Mallernee: I tell them the honest answer, which is that AI is going to change what family office professionals do, not how many of them there are. The routine mechanical work goes away: portal retrieval, document classification, transaction coding, reconciliation of the ninety-five percent of transactions that reconcile cleanly. That work should have been automated years ago and finally can be, and the professionals in the office get their weeks back.

What replaces that work is the analytical, judgment-based, principal-facing work that a family office is really there to do. The analyst is doing analysis. The controller is spending their time on the five percent of transactions that need judgment. The CIO is thinking about the two or three positions that really matter this quarter rather than reviewing the reconciled ledger. That is a better job. It also produces a better outcome for the family. The offices that have made this transition are not smaller. They are more capable, and their people are happier because the work is more interesting.

What comes next

Last question. If you had to name the single decision most family offices will get wrong in the next two years, what would it be?

Rob Mallernee: Waiting. Waiting for the technology to mature further. Waiting for their peers to move first. Waiting for the perfect moment. The technology is here. The reference operators are here. The governance framework is here. Every quarter that passes with an office still on legacy technology or spreadsheet operations is a quarter of compounding disadvantage relative to peers who have moved. Twelve months from now, the gap between AI-native family offices and everyone else is going to be visible to next-generation principals, to potential hires, to the families themselves. Now is the moment to move.

Frequently asked questions

Q: Who is Robert E. Mallernee?

A: Robert E. Mallernee is the Founder and Chief Executive Officer of Eton Solutions and a member of the Forbes Finance Council. He writes and speaks regularly on the architecture of wealth management technology and the operating model of the AI-era family office. Under his leadership, Eton Solutions has grown into one of the largest family office platforms in the world by assets on platform.

Q: What is an AI-native family office?

A: An AI-native family office is one whose platform was architected from the ground up assuming AI would do most of the routine operational work, rather than a platform designed for human operators with AI added on top. AtlasFive® by Eton Solutions is an example of an AI-native platform, with an integrated data model, knowledge graph, autonomous agents, and integrated governance framework.

Q: What is the difference between an AI-native platform and an AI-enabled platform?

A: An AI-native platform is architected from the beginning around AI performing autonomous work. An AI-enabled platform has added AI features on top of an architecture designed for human operators. The difference matters because AI-native platforms have integrated data models, knowledge graphs, and governance frameworks that AI can operate against natively. AI-enabled platforms have to work around the constraints of their original architecture, which limits what the AI can reliably do.

Q: Why does Eton Solutions emphasise ISO 42001 certification?

A: Eton Solutions emphasises ISO 42001 certification because it is the international standard specifically designed for how AI is governed, and it addresses the questions family office boards are now asking about their technology partners. As one of the first family office platforms globally to hold the certification, Eton believes ISO 42001 will become an expected qualification in family office platform RFPs over the next 18 months.

Q: Will AI replace family office professionals?

A: No. AI will change what family office professionals do, not how many of them there are. In an AI-native family office running AtlasFive® + EtonAI™, routine mechanical work is handled by autonomous agents, and professionals focus on analysis, judgment, and principal-facing work. The offices that have made this transition are more capable, not smaller. Their people also tend to be happier because the work is more interesting.

Q: What should family offices ask when evaluating an AI-enabled platform?

A: Family offices evaluating an AI-enabled platform should ask four specific questions. First, does the platform hold an authoritative record or read from systems held elsewhere? Second, does the AI understand entities and business rules, or primarily text? Third, can the AI execute within the system of record, or only produce answers? Fourth, is the governance independently certified against an international standard such as ISO 42001, or asserted in marketing materials? The answers separate platforms that will remain relevant in 2030 from those that will not.