Weather You Can Price: Planette's Joro Brings Long-Range Probability Forecasts to Insurance and Finance
Planette AI's Joro gives insurers and funds percentile weather outlooks, weekly to seasonal, worldwide. Distributions beat point forecasts for pricing, hedging and capital.

AI Weather Forecasts: The Next Edge for Insurance and Finance
Planette AI has launched Joro, a probabilistic weather intelligence platform built for insurers and financial institutions that need credible long-range forecasts. Instead of a single-point estimate, Joro delivers probability distributions for temperature and precipitation from the 5th to the 95th percentile, globally.
The output spans weekly outlooks up to six weeks and seasonal guidance across multiple months. For pricing, exposure management and market moves, that range matters more than a single number.
What makes Joro different
Joro blends physics-based climate models with AI, ingesting atmospheric, oceanic and terrestrial signals. It aggregates forecasts from multiple sources, including the National Oceanic and Atmospheric Administration (NOAA), the European Centre for Medium-Range Weather Forecasts (ECMWF) and Planette's own AI models, then applies performance-weighted algorithms to improve accuracy.
"Traditional forecasts might tell you it'll be a high temperature of 80 degrees, but that's just one outcome," says Dr Kalai Ramea, Co-Founder and CTO of Planette. "Our customers need to understand the full range of possibilities, including the likelihood of reaching 95 degrees."
Planette is also collaborating with NASA on "QubitCast," a quantum-inspired system that could extend predictive skill toward the six-month horizon. If successful, that would give risk teams more time to stage resources and capital.
NASA and ECMWF provide useful context on methodologies and benchmarks.
Why this matters for insurers
Rising claim costs, policy non-renewals and retreat from high-risk geographies reflect one thing: uncertainty. "Insurance companies are raising rates, declining renewals and pulling out of entire markets because they can't predict risk with existing tools," says Dr Hansi Singh, Co-Founder and CEO of Planette.
- Underwriting triage: Use tail probabilities to screen new business in heat, flood and wind-exposed ZIP codes.
- Pricing and filings: Adjust rating factors with percentile-based signals; document methodology for regulators.
- Reinsurance strategy: Set attachment points and buy limits using scenario distributions, not averages.
- Parametric products: Calibrate triggers to 95th percentile thresholds for rainfall, temperature or wind.
- Cat readiness: Pre-position adjusters and parts when probability spikes indicate surge risk.
- Capital and reserving: Feed distributions into VaR/TVaR, Solvency and IFRS 17 models to justify buffers.
Finance use cases that move P&L
Joro targets commodity desks and hedge funds that price weather into trades and risk. The tool surfaces low-probability, high-impact events that traditional forecasts miss, enabling better positioning and hedging.
- Energy: Anticipate heat-driven load, gas storage draws and power prices; stress-test generation and FX-linked exposures.
- Agriculture: Map rainfall and heat percentiles to crop yields, shipping and fertilizer demand.
- Infrastructure and real assets: Adjust maintenance windows, flood risk, and insurance costs in models.
- Portfolio risk: Use localized probability grids to refine scenario analysis and tail risk limits.
How it works under the hood
The platform blends multiple forecast sources and weights them by historical skill. Probabilistic outputs let teams model the full distribution, not just the mean. That improves decisions where tails dominate losses and slippage.
For diligence, ask about calibration and backtesting. Metrics like CRPS, Brier score and reliability diagrams matter more than headline accuracy.
Implementation checklist
- Data access: Joro is available via API for enterprise customers; confirm latency, uptime and SLAs.
- Coverage: Verify spatial resolution, elevation handling and sector-specific indices (HDD/CDD, SPI, heat index).
- Validation: Request out-of-sample performance by region and season; compare against internal loss data.
- Controls: Ensure audit trails, versioning and clear licensing for model outputs in filings and reports.
- Workflow fit: Integrate into pricing engines, catastrophe models, ETRM systems and treasury dashboards.
Where Joro fits in Planette's stack
Joro sits alongside Sura for general business applications and Umi for El Niño prediction. Planette also offers Eddy, a free public forecasting tool with customizable alerts.
Key quotes
"Our customers need to understand the full range of possibilities, including the likelihood of reaching 95 degrees." - Dr Kalai Ramea, Co-Founder and CTO
"Insurance companies are raising rates, declining renewals and pulling out of entire markets because they can't predict risk with existing tools." - Dr Hansi Singh, Co-Founder and CEO
Bottom line for risk leaders
If your pricing, hedging or capital process still runs on single-point weather inputs, you are flying blind in the tails. Probabilistic forecasts like Joro's give you the distribution you actually underwrite and trade against.
Test it against your own loss and P&L history. If the tails line up, you've found signal worth paying for.
Looking to upskill your team on practical AI tools for market and risk work? Explore our curated list of AI tools for finance.