CloudX is entering the market at a moment when mobile monetization is under real strain. Founded by Jim Payne, former CEO and co-founder of MoPub, and the original MoPub and MAX leadership team, the company launched publicly with a $30 million Series A to do something few ad tech platforms attempt: rebuild mobile monetization from first principles for an AI-driven world. Rather than layering automation on top of legacy mediation, CloudX is introducing an AI-native supply-side platform built around agentic systems, verifiable transparency, and secure auctions.

In this interview, Jim explains why investors believe this is the first true infrastructure shift since programmatic, why publishers are ready for autonomous monetization agents, and how CloudX’s launch signals a broader reset in how mobile advertising will operate in the years ahead.

 

Why do you think now is the right moment for investors to back agentic AI in mobile monetization? What industry shifts made CloudX’s approach feel urgent rather than optional?

We’re at the first real architectural shift in mobile monetization since programmatic arrived. The market has become far too large and too complex for static rules, manual workflows, or waterfall-era infrastructure. At the same time, publishers are dealing with privacy changes, signal loss, and rising user acquisition costs, which means every monetization decision has a greater impact than ever.

Agentic AI makes this complexity manageable. It finally allows publishers to operate the way they’ve always wanted to: segmenting audiences intelligently, setting dynamic price floors, structuring custom auctions, and reacting instantly to changes in demand. That level of sophistication used to be too costly or operationally impossible. Now it’s achievable if the platform has an AI-native design to leverage new intelligence and workflow agent capabilities.

So, the urgency comes from the market itself. Publishers need a fundamental upgrade in automation and transparency, not incrementally better mediation. Investors see that shift clearly.

Your Series A was significantly oversubscribed - what convinced investors that CloudX represents a fundamental new infrastructure layer rather than another optimization tool?

Most “optimization tools” live on top of existing systems. CloudX is the system. Investors understood that we rebuilt the entire monetization stack, the auction layer, the pricing layer, the creative controls, and the analytics as infrastructure designed to be operated by AI agents. Two things stood out to them:

TEE-secure auctions that provide verifiable fairness and protect buyer and publisher data. That’s not incremental; it’s a new trust model for mobile advertising.

Monetization as Code, where all monetization logic is expressed as files, not UI clicks. That unlocks true agentic automation and makes AI a first-class operator of the platform.

What they funded wasn’t an optimization feature. It was a new underlying architecture for how mobile monetization is going to work in an AI-native ecosystem.

How are publishers responding to the idea of fully autonomous monetization agents? Has adoption been faster or slower than you initially expected?

Faster actually, but with a healthy amount of curiosity and diligence. Publishers aren’t looking for a black box. What they want is automation with transparency: the ability to audit decisions, inspect logs, and configure the strategy to match their business.

The moment they see a TEE-secured auction, real-time log access, and the agent editing their monetization files exactly the way a human operator would, the hesitation goes away. They understand that this isn’t an autonomous system replacing human judgment; it’s a system that scales their judgment.

So adoption within the launch group has been strong because people immediately recognise how much manual work it offloads and how much sophistication it unlocks. We expect that to continue as the broader launch unfolds.

What pain points are publishers describing today that make agentic monetization particularly compelling as a replacement for traditional mediation and manual ops workflows?

Publishers mostly describe these recurring issues:

1. Manual setup doesn’t scale. Line items, floors, segmentation, anomaly detection, it’s too much surface area for teams to manage manually.
2. Lack of transparency. They want to understand why an auction cleared at a certain price or why certain buyers are outperforming.
3. As publishers’ businesses have grown, they need far more sophistication to optimize long-term user value, and static systems simply don’t support that.

Agentic monetization directly addresses all of these by allowing the platform to continuously adjust, learn, and optimize, while giving publishers full visibility into what’s happening under the hood.

How did your experiences building MoPub and MAX shape CloudX’s product architecture and what you felt needed to be “rebuilt for the AI era”?

MoPub and MAX were both built for the constraints of their time. MoPub introduced the first programmatic tools. MAX moved the industry toward unified auctions and liquidity. Both systems were built around UI-driven workflows and significant manual setup, which made sense at the time, but doesn’t scale in an AI-driven market.

With CloudX, we started by asking: What does a platform look like when the primary operator is an AI agent rather than a person?

That required rebuilding everything:

• Auctions secured by TEEs
• Monetization expressed as code
• Full log transparency
• An architecture where agents can set floors, create line items, detect anomalies, and deploy strategies in real time
• A data substrate designed for deep learning models

CloudX is what MoPub or MAX would look like if you built them now, with knowledge of the ecosystem’s scale, data complexity, and AI capabilities.

Why did you choose to build CloudX on Anthropic’s Claude technology, and what specific capabilities does it unlock for automated pricing and revenue optimisation?

Claude is extremely good at structured reasoning, code manipulation, and multi-step decision-making, which align perfectly with how monetization actually works.

Because our system is file-based, agents need to read, edit, and deploy monetization logic the same way an engineer would. Claude excels at this. It understands instructions, constraints, and policies, and it can reason across logs, line items, and historical performance.

Using Claude allows us to:

• Automate pricing strategies safely
• Detect anomalies in real time
• Run multiple simulations before deploying a change
• Ensure modifications are auditable and reversible
• Maintain a high level of consistency and governance

In short, Claude gives us agents that operate like highly skilled monetization engineers, but at machine speed and scale.

What are investors expecting CloudX to achieve over the next 12–18 months now that the platform is capitalized to scale globally?

Our focus in the next phase is on getting the platform into the hands of more publishers and expanding the ecosystem around it. That means:

• Bringing more publishers onto the SDK and into the secure auction
• Expanding liquidity by adding more demand partners over time
• Demonstrating that agentic automation can drive more sophisticated, long-term monetization outcomes for publishers

As more data flows through the system, the platform gets smarter. That’s when the flywheel really starts to spin, where insights from the auction feed directly back into agents that help publishers price, segment, and allocate inventory more effectively.

The goal is to build a foundation that can support much more automation over time, while keeping publishers in control and giving them full transparency into how decisions are made.

Do you believe agentic systems will change the role of monetization teams inside studios and publishers? If so, what does that future operating model look like?

Yes, but in a positive way. The goal isn’t to remove people from the process, it’s to take over the manual work that monetization teams spend a lot of time on today, such as setting floors, monitoring anomalies, tuning waterfalls, and fighting dashboards. 

With agentic systems, that work becomes automated. The role shifts toward:

• Defining business strategy
• Setting guardrails and policies
• Evaluating creative and user experience tradeoffs
• Using data to make higher-level decisions
• Coordinating across UA, product, and finance

Publishers are running much larger businesses now, and they need more sophistication to optimize them. Agentic systems make that level of sophistication practical, without adding operational overhead.