Finance Newsletter Writer Builds Custom AI News Scanner, Signals Shift in How Markets Process Information
A finance newsletter writer with no coding experience in four decades just built his own AI-powered news curation tool in a matter of days—a small but telling indicator of how rapidly AI is moving from engineering departments into the hands of business professionals who need to process vast amounts of market information.
Marc Rubinstein, who publishes Net Interest, a newsletter for finance professionals, used Anthropic's Claude Code to create software that indexes his five years of back issues and scans financial news daily to flag relevant stories. The system learned his editorial patterns from 250-plus archived newsletters, effectively creating a personalized research assistant without requiring traditional programming skills. For CFOs and finance teams drowning in information flows, the experiment offers a preview of how AI tools might reshape research workflows—though with significant caveats about what these systems can and cannot do.
The development comes as financial data providers face mounting questions about their business models in an AI-driven world. At a February 2026 conference, Euronext CEO Stéphane Boujnah delivered a striking assessment of the industry's data gold rush: "For years, I was told Euronext is missing the data revolution. Data is the new oil. You are missing the data boat. As I said earlier, all of us are finding out that maybe we missed the data boat—that maybe this data boat was a Titanic boat, that we missed the Titanic boat."
Boujnah's comment captures a growing tension in financial markets. Exchanges and data vendors spent years building businesses around selling market data, betting that AI would make their information even more valuable. Instead, they're discovering that AI models trained on publicly available information may reduce demand for expensive proprietary feeds—or at least force a reckoning with how that data is priced and packaged.
Rubinstein's experiment highlights both the promise and limitations of current AI tools for finance professionals. His custom system successfully identifies relevant news by learning from his historical coverage patterns. But it cannot access private information repositories, draw on personal networks, or replicate his writing voice—the elements that still require human judgment and relationships.
The adoption curve remains steep even among AI's active users. According to Anthropic, the median Claude Code session lasts just 45 seconds, and only 0.1% of users engage with the tool for longer than 40 minutes. The total active user base may number no more than one million. In a survey of 150 quantitative analysts and researchers in finance, adoption appears even more limited, suggesting regulated industries are moving cautiously.
For finance leaders, the implications extend beyond individual productivity tools. If professionals with no recent coding experience can build custom AI applications in days, the traditional boundaries between business users and technical developers are eroding faster than most organizations have prepared for. That shift raises questions about data governance, model risk management, and how finance functions should structure their technology capabilities.
The more immediate question may be what happens to the financial information industry itself. If AI can effectively curate and synthesize publicly available market data, the premium that exchanges and data vendors have commanded for decades faces pressure. The "data is the new oil" narrative that drove valuations and strategy may need revision—or as Boujnah suggested, the boat everyone was chasing may have been headed for an iceberg all along.
Responses (0 )