---
title: privacy-policy
---

[![Betty\_Eggplant\_TasioTeal](https://meetbetty.ai/hubfs/eggplant+tasioteal@2x.png)](https://meetbetty.ai/?hsLang=en)

- [Product](https://meetbetty.ai/product)
- [Case Studies](https://meetbetty.ai/case-studies)
- [Betty Blog](https://meetbetty.ai/bettys-blog)
- [Pricing](https://meetbetty.ai/pricing)
- [About](https://meetbetty.ai/about)

[Book a Demo](https://meetbetty.ai/demo-1?hsLang=en)

# **Environmental Impact & Sustainability**

Last updated: October 6, 2026

Blue Cypress is a family of purpose-driven companies dedicated to empowering associations, nonprofits, and the broader social sector with AI. As proud practitioners of Conscious Capitalism, we believe that when technology uplifts mission-driven organizations, it creates a multiplier effect of good in the world. That belief shapes how we build and operate our software — including our responsibility for its environmental footprint.

**We don't train our own models — we build small, purpose-built AI.**

Blue Cypress does not develop or train its own AI models. That matters environmentally: training large models is by far the most energy- and water-intensive part of the AI lifecycle — a cost we never incur. Instead, Betty, Skip, rasa.io, Izzy, and our other products are narrow, task-specific assistants that draw on existing models. Betty, for example, answers members' questions using an association's own vetted content — a retrieval-grounded design that calls a model briefly for each answer rather than training or operating a massive general-purpose model. Purpose-built tools like ours consume only a small fraction of the energy and water associated with developing and running large frontier AI models.

**We're not locked in to any vendor — and we treat model choice as an environmental feature.**

A core value of Blue Cypress products is the freedom to adapt as AI evolves and to let each organization use the models it prefers; we are not locked in to any single AI vendor. Our preferred model vendors today are Google Gemini and Cerebras (running efficient open-source models), chosen for speed, cost, and efficiency, and we can move to others as the field changes. Because we are model-agnostic, customers are not locked into the largest, most energy-intensive systems: they can favor right-sized and open-source models, and weigh environmental cost alongside quality, speed, privacy, and price. While we can't put a precise number on any single model's footprint, we can work with customers who have a preferred provider — or specific providers they would rather we avoid, including on environmental grounds.

**We run on shared, hyperscale cloud infrastructure — by design.**

Today, our products run on Microsoft Azure's shared, multi-tenant cloud rather than dedicated, single-tenant data centers, and some are beginning to adopt Amazon Web Services (AWS) as well — MemberJunction Central, for example, uses AWS as its preferred setup path. Shared infrastructure is materially more efficient: compute, power, and cooling are pooled across many organizations and run at high utilization, and hyperscale facilities operate far more efficiently than typical private or on-premises data centers. Google reports that its data centers use about 84% less overhead energy than the industry average, and AWS reports a power-usage effectiveness of about 1.15, well below typical on-premises data centers.

**We choose providers with credible environmental commitments.**

Where we rely on third parties — cloud platforms like Microsoft Azure and AWS for hosting, and AI models from providers such as Google and Cerebras — we favor partners who publish real environmental commitments and disclose their progress. Microsoft has committed to be carbon negative and water positive by 2030, reports that it met its goal of matching 100% of its electricity use with renewable-energy purchases, and has begun deploying next-generation data-center designs that consume zero water for cooling. Google reports industry-leading data-center efficiency and a goal to replenish more freshwater than it consumes by 2030. Amazon Web Services has likewise committed to be water positive by 2030 and net-zero carbon by 2040, reports matching 100% of its electricity use with renewable energy in 2023, and cools its data centers with outside air roughly 90% of the time — using water only on the hottest days.

**We are honest about what we can — and cannot — precisely measure.**

The candid reality is that the exact environmental footprint of a single organization's usage, or a single AI query, is difficult to quantify with precision today. We operate on shared infrastructure where impact is pooled across millions of users, and the industry does not yet have standardized, independently audited per-query metrics. Rather than publish a falsely precise number, we commit to transparency about how our systems work and to sharing the published data of the providers we depend on.

**Our commitments**

- **Efficiency by design —** keep building right-sized, retrieval-grounded products that minimize unnecessary computation.
- **Responsible provider selection —** prefer cloud and model providers with credible, published commitments to carbon-free energy and water stewardship, and monitor their progress.
- **Flexibility and choice —** stay model-agnostic so we — and our customers — can shift to more efficient models over time, and so environmental cost can be part of model selection.
- **Transparency —** give customers clear, accurate information, including the limits of what can be measured, and point to primary sources.
- **Continuous improvement —** as measurement, models, and hardware become more efficient, adopt them and revisit this position. Progress over perfection.

*We hold ourselves to the same standard we bring to everything else: do right by the organizations we serve, be honest about what we know, and keep getting better.*

[![Betty\_Eggplant\_TasioTeal](https://meetbetty.ai/hubfs/eggplant+tasioteal@2x.png)](https://meetbetty.ai/?hsLang=en)

AI knowledge system built exclusively for associations. Betty does the jobs your members and staff need.

[support@meetbetty.ai](mailto:support@meetbetty.ai)  
[+1 (202) 935-9122](tel:+12029359122)

##### Product

- [How Betty Works](https://meetbetty.ai/product)
- [Pricing](https://meetbetty.ai/pricing)
- [Referral Program](https://meetbetty.ai/refer-betty)
- [Book a Demo](https://meetbetty.ai/demo)

##### Company

- [About](https://meetbetty.ai/about)
- [Partners](https://meetbetty.ai/partners)
- [Speaking & Media](https://meetbetty.ai/speaking-and-media)
- [Contact Us](mailto:%20support@meetbetty.ai)

##### Resources

- [Case Studies](https://meetbetty.ai/case-studies)
- [Betty Blog](https://meetbetty.ai/bettys-blog)
- [Content Readiness Checklist](https://meetbetty.ai/ai-ready-content)
- [AI Readiness Checklist](https://meetbetty.ai/ai-assistant-readiness-checklist)
- [Build a Prompt Library](https://meetbetty.ai/build-a-prompt-library)
- [AI Implementation Roadmap](https://meetbetty.ai/ai-implementation-roadmap-for-associations)

© 2026 Betty. All rights reserved.

[Privacy Policy](https://meetbetty.ai/privacy-policy-1?hsLang=en) [Terms of Use](https://meetbetty.ai/terms-1?hsLang=en) [Responsible AI](https://meetbetty.ai/responsible-ai-1?hsLang=en)

```json
{
  "@context" : "https://schema.org",
  "@type" : "Organization",
  "alternateName" : "Betty AI",
  "contactPoint" : {
    "@type" : "ContactPoint",
    "contactType" : "customer support",
    "email" : "support@meetbetty.ai",
    "telephone" : "+1-202-935-9122"
  },
  "description" : "AI knowledge platform built exclusively for professional and trade associations.",
  "logo" : "https://meetbetty.ai/hubfs/Betty%20Defaults/BettyLogo.svg",
  "name" : "Betty",
  "sameAs" : [ "https://www.linkedin.com/company/meetbetty", "https://www.crunchbase.com/organization/betty-ai" ],
  "url" : "https://meetbetty.ai"
}
```