
The claim — circulated without attached methodology, benchmark thresholds, or execution-assumption set — lands in a market where the infrastructure for that displacement is already measurable.
The execution layer is going public
Concrete data is arriving in the same window. Crypto Briefing reports that Gemini launched a public REST API on July 30, serving prediction market trading volumes by category, broken down by day or hour in UTC. No authentication. No API key. An unauthenticated endpoint compresses the friction cost of backtesting volume-derived signals to near zero — any HTTP client can now ingest the series.
Gemini Predictions, operating as a CFTC-approved Designated Contract Market since its December 16, 2025 launch, had crossed $24 million in cumulative trading volume and 10,000 active users by July 2026. Q1 2026 contribution to exchange revenue: approximately $400,000 against a $50.3 million total (+42% YoY). Cumulative contracts traded since launch: over 100 million. Batch-order APIs shipped July 17. Event categories were extended to include FIFA World Cup contracts in the same update cycle.
Sector volume is the variable
The market is no longer fringe. Cryptopolitan data places July 2026 prediction market volume at a record $55.5 billion, with Kalshi accounting for roughly 70% of total trade count. April 2026 alone delivered a 78% month-over-month volume jump at Gemini.
Three data points to track against the 2028 horizon:
- API surface area. Each new unauthenticated endpoint widens the addressable dataset for latency arbitrage and signal generation. Monitor release cadence across venues.
- Volume concentration. Kalshi's 70% share is a single-point-of-failure risk for any strategy benchmarked against the sector. Track the HHI of trade count monthly.
- Contract throughput. 100 million contracts across one venue in seven months sets the throughput reference. Cross-reference against execution latency benchmarks from the same period.
Risk-adjusted verdict
The CEO's statement is a directional claim, not a model output. No Sharpe ratio, no slippage parameter, no drawdown constraint. Treat it as a sentiment indicator from a single counter — and let the public volume APIs perform the verification. The infrastructure is now accessible enough for any quant to backtest the claim against execution-grade data within a single working session.