⚡ Try it now — Ask your AI how to connect ActuaryMCP: https://mcp.actuary.meridianbridgegroup.com/mcp

Your AI Is Guessing Actuarial Rates.
Fix That.

ActuaryMCP connects ChatGPT, Claude, and Copilot to 3,000+ SOA mortality tables, NAIC PBR spreads, and FRED interest rates. No hallucination. Just real data.

Connect Your AI See It in Action
3,000+
SOA Mortality Tables
20
MCP Tools
4
Data Sources
$0
Cost to Connect

The Problem

Without real data, AI tools hallucinate actuarial rates. With ActuaryMCP, they pull verified values directly from authoritative sources.

Gen AI — Without ActuaryMCP
What is q45 for a 45-year-old male, 2017 CSO ANB unloaded nonsmoker?
The 2017 CSO ANB table for a 45-year-old male, unloaded:

q45 = 0.001720 (1.720 per 1,000)

A few notes:
• ANB = Age Nearest Birthday
• Unloaded means the pure mortality rate without any loading margin
• The 2017 CSO replaced the 2001 CSO and generally shows lower mortality rates
• This is the male non-smoker composite rate
⚠ Actual q45 Super Preferred select = 0.00025 — this answer is off by 6.9x
Is this real data? Can you verify it?
I don't have direct access to the SOA mortality table files. The value I provided is based on general knowledge from my training data and may not be exact.

For verified values, I'd recommend downloading the official 2017 CSO tables from the SOA website or consulting your company's actuarial library.
Gen AI — With ActuaryMCP
What is q45 for a 45-year-old male, 2017 CSO ANB unloaded nonsmoker?
⚙ search_tables("2017 Unloaded CSO Preferred NS Male ANB")
⚙ lookup_rate(table_id=3341, age=45, duration=1)
Here is the full q45 picture for all three 2017 Unloaded CSO ANB Male nonsmoker preferred risk classes:
Risk Class Select (Dur. 1) Ultimate
Super Preferred 0.00025 0.00128
Preferred 0.00034 0.00151
Residual Standard 0.00042 0.00172
  • Select rates span ~1.7x from Super Preferred to Residual Standard, reflecting underwriting differentiation
  • Ultimate rates converge — the select underwriting credit fades as duration extends
  • Select-to-ultimate ratio is roughly 5x across all three classes
Source: SOA Tables 3341, 3343, 3345 — verified via ActuaryMCP

What It Does

One MCP server connecting your AI to four authoritative actuarial data sources.

📊

SOA Mortality Tables

3,000+ tables including 2017 CSO, VBT, select-and-ultimate structures. Search by name, look up rates by age and duration, get full metadata.

📈

NAIC PBR Spreads

Tables A, F, and G default costs and spreads by credit rating and weighted average life. Pulled directly from NAIC XLSX files.

🏦

FRED Interest Rates

Treasury yields, swap rates, and economic indicators in real time from the Federal Reserve Economic Data API.

🔄

Rating Conversions

Map seamlessly between Moody's, S&P, Fitch, and NAIC credit rating designations. No more lookup tables in spreadsheets.

Connect in 30 Seconds

Works with any AI tool that supports the Model Context Protocol.

1

Copy the MCP URL

One URL connects your AI to all 20 actuarial tools.

https://mcp.actuary.meridianbridgegroup.com/mcp
2

Paste in Your AI Settings

Add as an MCP server in ChatGPT, Claude Desktop, GitHub Copilot, or any MCP-compatible tool.

3

Ask Any Actuarial Question

Your AI now pulls verified data from SOA, NAIC, and FRED instead of guessing.

Stop Guessing. Start Computing.

Free to use. No data collected other than function call counts. Built and verified by an actuary (FSA, MAAA).

Connect Your AI