The blog →
Fact-checking with AI is starting to work
We're releasing Big If True, a free Claude skill that works through a text one claim at a time, verifies each one against sources, and shows its receipts.
September 16, 2026
We build AI tools for newsrooms, journalists, and creators that push what's possible. We ship fast, turning the latest research into working products.
We teach teams to work with AI rather than just talk about it. Our workshops, hackathons, and coaching leave participants with tools and skills they can use right away.
The AI landscape changes every week. We help leadership teams cut through the noise and make smart decisions about what matters now and what to ignore.
A free Claude skill that works through a text claim by claim, verifies each one against sources, and hands back a report with receipts. It caught real errors in real published stories and shows that fact-checking with AI is starting to work.
Too many podcasts, not enough time. Re:verb listens to every new episode of your favorite shows and sends you a quick summary. Roganbot does the same for the world's biggest podcast, pulling key moments and quotes out of every Joe Rogan episode.
Monitors police radio around the clock so local newsrooms don't have to. Flags newsworthy events in real time and transcribes every call for follow-up. Currently piloting with Open Vallejo in California.
Virtual focus groups, built from real interview data. Test a headline, a message, or a product idea against a realistic model of your actual audience, and get an answer in minutes rather than months.
A half-day workshop to get your team up to speed with AI. We cover the fundamentals, run live demos, and make sure everyone leaves with tools they can start using right away.
One-day sprints where your team builds a working AI prototype from scratch using tools like Claude Code. Show up with a problem, leave with something you can bring to your newsroom.
Multi-month programs that go beyond the basics. We run the fjum AI Media Academy in Vienna, and just launched the WAN-IFRA NextGen AI Leaders Programme.
Support for teams and individual builders, so you're not going it alone. We work alongside your team as you develop projects, with regular check-ins from first idea through to launch.
We serve on boards focused on the forces reshaping media: AI, new business models, and the changing relationship between creators and audiences.
Focused engagements with your leadership team to separate hype from reality and map out an AI plan for the coming months and years, ending in concrete next steps.
The view from San Francisco. We bring the frontier of AI development to your organization or conference, grounded in real-world experience building and shipping AI products.
Guidelines and implementation frameworks for responsible use of AI in your organization, offering clear, practical policies your team can actually follow.
A free fact-checking skill for Claude. It checks your draft one claim at a time and shows you what needs a second look, with sources.
Verso was founded by Patrick Swanson and Kaveh Waddell as JSK fellows at Stanford University. We're journalists who understand AI, not a tech startup trying to understand media.
Patrick has a decade of leadership experience in media and tech. He led the social media team at Austria's largest newsroom, reaching 2.4 million daily across TikTok, Instagram, and Facebook.
Kaveh is an investigative data journalist who led technology investigations at Consumer Reports, launched the AI beat at Axios, and covered surveillance for The Atlantic.