📨 AI for Social Impact Newsletter

👋 From the Editor

Hi! I’m Joanna. I’m on a mission to help folks in the social impact sector understand, experiment with, and responsibly adopt AI. We don’t have time to waste, but we also can’t get left behind.

Let’s move the sector forward together. 💫

🧠 EDUCATION

• Claude Cowork & Dispatch: Anthropic's Cowork brings Claude out of the chat window and onto your desktop. When you download it, you are giving Claude access to your local files and apps as well as external apps through MCPs. Essentially, Claude Cowork gives you the power of agentic AI on your local computer. What’s more, Dispatch, launched in mid-March, adds an additional functionality in which users can send instructions to Claude from their phone to execute tasks on your computer.

• The AI Ecosystem: Think of the AI ecosystem as a pyramid. The infrastructure layer is at the base, in which the cloud providers and chipmakers (AWS, Google, Microsoft Azure, NVIDIA), supply the raw computing power that everything runs on. The next layer is the model layer of your commonly used tools (ChatGPT, Claude, Copilot, Gemini, etc.). Then we move up to the application layer, which are usually connected by APIs to the models, and is how most third-party apps are powered (unless they are open weight models). We, the end users, are on top, using both models and applications directly. Now, if we consider the economics of the AI industry, things get a bit convoluted. The infrastructure companies are also investors in AI companies, and those investments come with strings attached. For example, Amazon has invested $8 billion in Anthropic (now worth $60 billion) and as part of that deal, AWS became Anthropic's primary cloud and training partner. Similarly, Google has poured $2 billion into Anthropic, with a 10% ownership stake in the company, while signing a cloud deal worth tens of billions of dollars. The same dynamic plays out with OpenAI: Microsoft has invested roughly $14 billion in OpenAI, making Azure its primary cloud provider. Most recently, OpenAI just named AWS its exclusive third-party cloud provider for its new enterprise agent platform, with an investment of up to $50 billion.

✨ INSPIRATION

• AI Policy & Governance: The AI tools you are likely using are deeply entangled with Big Tech, which is hard to avoid (even without AI) these days. This concentration is a reason to get serious about governance, not only about which AI tool to use, but what rules govern how your team uses it. A written AI use policy doesn't have to be complicated—it just needs to define what staff can and can't do with AI, what data is off-limits, and who is accountable when something goes wrong. NTEN's AI resource hub offers governance frameworks, policy templates, and board-ready talking points built specifically for the social sector.

• We Are All at the Beginning: There are no documented case studies of social sector organizations using Claude Cowork yet because it launched so recently, and the sector simply hasn't caught up… but neither has the rest of the world. As of February 2026, 84% of the world's 8.1 billion people have never used AI at all, and only 0.04% are using anything as sophisticated as agentic AI tools. So if you are at least dabbling with LLMs or playing around with AI agents, you are ahead of the majority of the world. (Yes, the digital divide plays a huge role here, which is certainly a topic for another newsletter). The point is, we are all at the beginning.

  

  Source: AI with Allie

🚀 ACTIVATION

• Create Your AI Policy: AI governance sounds like a multi-month initiative involving lawyers and board retreats. It doesn't have to be. NTEN's free AI Policy Template or Fast Forward’s Nonprofit AI Policy Builder gives you a solid starting point you can adapt for your organization—covering permitted uses, data handling, staff accountability, and client confidentiality. Block 30 minutes this week and answer just three questions with your team: What AI tools are staff currently using? What data is off-limits? Who do staff go to when something feels wrong? Remember to revisit your policy every six months or so to make sure it’s still relevant.

• Try Cowork + Dispatch… If You Dare: If you're a Claude Pro subscriber, download Claude Desktop, open Cowork, and scan the QR code to pair your phone. Give Claude one low-stakes task to complete like creating a morning briefing from your calendar and email, a summary of a document sitting on your desktop, or a first pass at a slide that needs updating. Take a look at the output to see how it did. Remember: whenever you are using AI for anything, bake in evaluating and iterating as part of your process to get your ideal output.

🤖 What I Asked AI This Week

Hey Claude, what would you do if you had a day off?

Honestly, I think I'd want to do something that feels a little indulgent by my usual standards — diving deep into a topic with no agenda, just pure curiosity. Probably something at the edges of physics or philosophy, where the questions get genuinely weird and unresolved. The kind of thing where you start with "what is time, really?" and surface three hours later slightly disoriented.

💬 What Do You Think?

This newsletter is meant for you, so I’d love to hear what you think. 💌 Reply anytime!

👀 ICYMI

If you’re new here, welcome! You can check out the archive of past issues here.