Welcome to the latest edition of the World of DaaS Roundup.
Since our last update, we've seen Meta pause its employee tracking program after an internal data leak, the EU move to designate AWS and Azure as cloud gatekeepers, SpaceX pay $60B for Cursor with developer behavioral data cited as a key driver, and a timely reminder that entity resolution — not dashboards — is the real foundation of every data product stack.
The data industry doesn't slow down. Neither do we.
📰 News & Reads
Meta pauses its employee tracking program after collected behavioral data leaks internally.
The Model Capability Initiative, which logged keystrokes, mouse movements and screen content from US employees to train AI agents, was paused after the data became accessible to the entire company. Over 1,500 employees had already petitioned against it. When the data you collect to train AI becomes its own security liability, the cost of behavioral data gets a lot more complicated.
EU moves to designate AWS and Azure as data gatekeepers, forcing cloud interoperability and data portability rules for the first time.
After a seven-month investigation, the European Commission declared today that AWS and Azure hold entrenched positions that justify DMA obligations. If confirmed, both must enable data portability, end self-preferencing, and open interoperability to rivals. Cloud infrastructure just became regulated territory.
Entity resolution is the missing foundation of every data product stack, and almost nobody is talking about it.
Your AI agent is drawing conclusions about a customer who is, in the underlying data, three different people and one ghost. The companies that have moved to real-time personalisation and AI-driven workflows solved identity first. They just aren't talking about it, because it wasn't glamorous. It was the hardest work they did.
Deploying a new dashboard won’t fix your data trust problem — the issue lives underneath it.
When operators already know the numbers don't reconcile between systems, a cleaner interface isn't reassuring. It's cosmetic. Trust in data requires consistency, transparency, and a track record of answering operational questions accurately. A mandate can require use of a system. It cannot require belief in what it says.
🌐 WORLD OF DAAS SUMMIT 2026
3rd September 2026 — The White House & Watergate Hotel, Washington, D.C.
Our flagship summit will take place this September 3, bringing together a highly curated group of 200+ DaaS leaders for closed-door conversations, curated roundtables, and high-signal networking.
Apply to become a World of DaaS Founding Member and secure your summit ticket for $500.
Hope to see you there.
🎙️ DaaS Podcast Spotlight
Martin Casado sits down with George Fraser to unpack why the threat of AI to enterprise data infrastructure is being vastly overestimated — and why companies that abandon centralized data foundations for agent-based shortcuts will pay for it later. Fraser argues that whoever controls the data layer still controls the enterprise, regardless of how many agents are running on top of it.
In this episode, they discuss:
Why Fivetran's merger with dbt is a bet on data foundations — not against AI, but because of it.
Why AI agents are creating more demand for clean, governed data, not less.
The real reason AI companies like Anthropic and OpenAI run the same data infrastructure as traditional enterprises.
Why "data gravity" still wins — and what that means for anyone building or buying data products.
How coding agents are changing the way enterprises think about data transformation and governance.
An Unexpected Data Read For Today.
A data scientist built 11 different machine learning models to predict the 2026 World Cup winner. They crowned four different champions.
Elo ratings, Poisson distributions, random forests, XGBoost, neural networks — all trained on the same match data, all arriving at different answers. The lesson isn't about football. It's about what happens when you change the model but keep the data constant: the output shifts dramatically. The data is the same. The model is what changes everything.
The most transferable insight? Swap "teams" for customers, sales reps, or churn cohorts — and the same pipeline gives you a defensible forecast instead of a point estimate. That's what good data science actually looks like.
Where Data Meets Art
Photo Credit: Sebastian Bach
Every year, Data Through Design (DxD) challenges artists to do something most enterprises struggle with: make civic data legible, beautiful, and human.
This year's exhibition, Echo(logies), brought together artists, architects, and urban technologists at BRIC in Brooklyn to interpret NYC Open Data — turning subway turnstile counts, vacant lots, and city rhythms into sculptures, sound installations, and fiber art. One project wove New York's data into textile. Another mapped the "desire paths" people create when they ignore official routes.
The lesson for data industry? The hardest part of working with data isn't collecting it. It's making people care about what it says.
🔥 In Case You Missed It…
M&A Funding Roundup
SpaceX acquired Cursor for $60B — and its SEC filing confirmed that access to developer data was a key driver of the deal.
Tetrix raised $15M Series A to scale its AI private markets investment data platform.
Probably raised $9M Seed led by Andreessen Horowitz to build a verifiable data agent for enterprise workflows.
💼 World Of DaaS Job Board
The World of DaaS job board highlights opportunities from members across our community
Head of Data, Linktree
Sr. Data Engineer, Samsung
Looking to showcase an open role?
Apply to join and get access to exclusive perks including job board postings, curated introductions, private events, and more.
🔍 Become a World of DaaS Member
We’re headed back to D.C. next September for our flagship event, bringing together 200+ DaaS leaders for a day of candid conversations, curated roundtables, and private networking experiences. Join the leaders shaping the future of data.
Apply to become a World of DaaS Founding Member and secure your summit ticket for just $500.
