Mozilla

Who really controls enterprise AI agents? Join @raffi live on CXO Talk to hear why open-source AI may be the key to independence, what closed platforms risk, and why visibility and control matter. cxotalk.com/episode/mozill…

AI adoption without trust is a losing strategy. Stephanie Mehta from Fast Company interviews Mark Surman on why companies need to focus on employee agency, meaningful AI governance, and accountability if they want people to trust how AI is being used. “The consequences of being untrustworthy and ignoring accountability are through the roof.” The companies that build trust alongside AI innovation will win. fastcompany.com/91543169/mozil…

The AI stack is reorganizing in real time. Raffi Krikorian’s newest newsletter connects everything from Karpathy’s “microgpt” experiment to on-device AI, federated learning, agent harnesses, and Minecraft being used to bypass censorship and preserve press freedom. Read here: newsletter.ownersnotrenters.com/p/the-complete…

What should AI users actually be allowed to see? In an interview with @TheAtlantic CEO @nxthompson , @raffi argues that open-source AI could make systems more understandable, accountable, and trustworthy — while also raising important questions about safety and control. youtube.com/watch?v=uK7Eca…

The next chapter of AI isn’t about who adopts it — it’s about who owns it. Mark Surman and Valérie Pisano make the case for open-source AI as the foundation for a more trustworthy, shared global ecosystem in their new op-ed. Not Big Tech vs. protectionism — but something better. theglobeandmail.com/business/comme…

@raffi's latest newsletter focuses less on AI itself and more on incentives. There’s no shortage of problems to solve — just a shortage of funded time to solve them. The companies betting on open models are already hiring into that gap. Subscribe here: newsletter.ownersnotrenters.com/p/lets-get-you…

@raffi's newest newsletter had this takeaway: AI progress isn’t really stuck on models anymore. It’s everything around them — permissions, memory, cost, agents in real systems — that still doesn’t fully exist yet. Read here: newsletter.ownersnotrenters.com/p/ai-agents-ro…

“Generic benchmarks are a tiebreaker, not a decision. The set that matters is the one you’re generating every day.” That’s the shift. @raffi is sharing the full system and tooling in his newsletter. If you’re using AI tools for real work, you shouldn’t be relying on benchmarks. You should be evaluating models on your own prompts and the workflows you actually run. Sign up here: newsletter.ownersnotrenters.com/p/save-the-pro…

“Now, thanks to new A.I. tools, anyone can write code… The détente is over.” @raffi, wrote an op-ed in @nytimes this week on how AI is reshaping cybersecurity. The balance between building and breaking software is disappearing—and it’s happening fast. Read it here: nytimes.com/2026/04/15/opi…

Open source AI often gets dismissed as “tinkering” — but it’s often where real infrastructure starts. No sales teams. No polished packaging. Just systems that become essential. Read the report by Startup Coalition here: api.startupcoalition.io/u/2026/04/Hidd…

What does it take to build a more open, people-centred internet? On CDT’s Tech Talks, Mark Surman joined Alexandra Reeve Givens to talk AI governance, privacy, platform accountability, and what it takes to build a more trustworthy digital future. Listen here: cdt.org/insights/talki…

Today we announced Thunderbolt, an AI client built on the same principles that have guided Mozilla for decades: user control, open standards, and individual empowerment online. As AI becomes central to enterprise operations, organizations face a critical choice: depend on proprietary services that create strategic risk, or build on open infrastructure they control. Learn more: thunderbolt.io/announcing-thu…

AI is starting to act as the layer between us and everything online—shaping what we see, decide, and create. Mozilla argues that if it’s all closed systems, we don’t really own that intelligence—we rent it. Tune in: blog.mozilla.org/en/mozilla/moz…

AI is reshaping the internet—but also concentrating power. Mozilla is pushing for AI that’s open, trustworthy, and human-first. Feels like a Web 2.0 moment again—who will control what we see and build? Here's our plan: blog.mozilla.org/en/mozilla/rew…

The State of Mozilla was published two months ago, but is even more relevant as the AI landscape keeps shifting. It lays out our roadmap for open, sovereign AI and the work happening across Firefox, Mozilla.ai, and our data + developer tools. stateof.mozilla.org

Raffi Krikorian joins Inference to break down a question a lot of teams are quietly wrestling with: when does open source AI stop being a nice idea and start being the practical choice? He talks real economics, the missing “LAMP stack for AI,” and why the hardest part isn’t models — it’s the glue that makes them usable. Catch the episode for a grounded take on switching from closed APIs to open models: turingpost.com/p/krikorianhtt…

Remember when Firefox broke Microsoft’s browser monopoly? Mozilla did that for the web. Now it’s AI’s turn. Owners Not Renters by Raffi Krikorian guides developers & creators to take back control from closed AI platforms. Don’t miss the first edition: newsletter.ownersnotrenters.com/p/all-about-op…

Mark Surman argues that today’s AI boom looks a lot like the dot‑com bubble — concentrated power, inflated valuations, and an economic model built on extraction, not public benefit. The real question is what comes after the crash. Read here: theguardian.com/commentisfree/…

Fast Company highlights Mozilla’s shift toward a different kind of AI future — one where people decide when AI shows up, not the other way around. New Firefox controls are just the start. The larger goal is a “rebel alliance” for open, sovereign AI: developers, researchers, and communities building technology that serves people, not platforms. Read here: fastcompany.com/91490055/mozil…