FiscalNote Holdings, an AI-driven policy intelligence firm, lost its NYSE listing on March 25 after failing to maintain a $1.00 average share price over 30 trading days. The company had already executed a reverse stock split within the past year, leaving no further cure period available.
The delisting caps a years-long decline that traces directly to LLMs commoditizing the policy data middleman model on which FiscalNote built its business.
A Policy Intelligence Pioneer Falls Below $1
The New York Stock Exchange announced it would commence delisting proceedings against FiscalNote (NOTE) under Section 802.01C of its Listed Company Manual. Trading in Class A common stock and warrants was suspended immediately. Shares are expected to begin trading on the OTC market on March 26 under the same ticker.
The timing is striking. On the same day, the NYSE moved to delist the company, and FiscalNote announced its PolicyNote MCP server had been approved for the OpenAI App Store. The listing gives ChatGPT’s 700 million weekly active users access to structured policy data across Congress, all 50 US states, and over 100 countries.
FiscalNote called the delisting the start of “a new phase of health and opportunity.” The company pointed to a 25% reduction in workforce and a 19% cut in cash operating costs. It projected positive free cash flow over the twelve months beginning April 2026.
Founded in 2013, FiscalNote sells legislative tracking and regulatory intelligence to enterprises and governments through its flagship PolicyNote platform. The company went public in 2021 via a SPAC merger.
The SaaSpocalypse Claims a Data Middleman
FiscalNote’s decline fits a pattern analysts have labeled the “SaaSpocalypse” — the structural erosion of SaaS business models by LLM-powered agents. Since early 2026, the broader software sector has shed roughly $1 trillion in market capitalization as enterprises shift budgets from per-seat subscriptions to agentic AI platforms.
FiscalNote built its moat on a specific form of information asymmetry. Policy documents are public, but collecting, structuring, and interpreting them at scale was expensive. LLMs collapsed that cost. Any compliance team can now feed a bill’s text into an AI model and extract a summary that once required a FiscalNote subscription.
The company’s pivot confirms this diagnosis. Launching a PolicyNote MCP server on March 4 signaled a retreat from selling analysis to selling raw data infrastructure. In effect, it conceded that LLMs handle the interpretation layer better and cheaper.
A String of Crypto-Adjacent Pivots
FiscalNote has tried multiple crypto-flavored strategies over the past year. In June 2025, it began evaluating stablecoins as a payment method for international customers. By September, the company announced it was exploring Bitcoin, Ethereum, and Solana as strategic treasury reserve assets, following the corporate BTC treasury trend popularized by MicroStrategy.
In February 2026, FiscalNote expanded into political prediction markets. It launched a preview at PoliticalPredictions.com, signed a non-binding MOU with 365Prediction, and appointed former Sportradar executive Dr. Laila Mintas as a strategic advisor. Last week, it signed an MOU with Korean law firm D&A LLC to distribute US policy intelligence across Asian markets.
Each move targets a different growth narrative. None has yet translated into revenue sufficient to keep the stock above $1.
Prediction Markets May Outlast the Company’s Other Bets
Of FiscalNote’s recent initiatives, the prediction-market push aligns most closely with where the broader market is headed. Monthly trading volumes in prediction markets now reach approximately $10 billion. Kalshi has overtaken Polymarket as the volume leader, capturing roughly 66% of market share.
Policy intelligence and prediction markets share a natural logic. Enterprises do not really need to know what a bill says. They need to know whether it will pass, and when. That is a probability question. Prediction markets are the native format for pricing probabilities.
But there is a structural tension. Prediction markets thrive on liquid, high-interest events — elections, rate decisions, wars. The niche regulatory questions where FiscalNote’s data has the most value — “Will the EU AI Act Article 6 high-risk classification criteria be finalized by Q3?” — are precisely the ones too obscure to attract betting volume.
FiscalNote’s board said it continues to review all strategic options, including potential divestitures of non-core assets. Whether the company can execute any of its pivots from the OTC market remains an open question. Less ambiguous is the lesson its trajectory illustrates: in the LLM era, companies that monetize the gap between public data and user comprehension face existential pressure. The middleman’s margin is compressing, and FiscalNote’s delisting is a data point in that trend.
