This demonstrates how managed AI services and serverless infrastructure lower the barrier to entry for complex application development, allowing small teams to build secure, scalable, and cost-effective tools without deep infrastructure expertise.
Tracking how governments target dissidents living in exile. Helping people in crisis find mental health support. Advocating for legislation that protects free expression online. These are a few examples of how some of the world's leading organizations are building the future of non-profit work with Cloudflare.
AI is changing how people do their work. The goal of Cloudflare Impact is to help ensure that non-profit organizations are among the first to benefit. Today, we’re sharing what dozens of civil society organizations have built using our developer services with more than $7.5 million of Cloudflare credits. These stories show what is possible when AI applications are accessible, secure, and affordable to build and run.
We believe a better Internet is one that allows people to express themselves online and access a diverse range of viewpoints. A key part of Cloudflare's mission has been making security services available for everyone and helping ensure that individuals and organizations working for the public interest are not forced offline by those more powerful. Today, Project Galileo, which provides free cybersecurity services to civil society organizations, protects more than 3,400 domains across more than 120 countries.
Through these partnerships, organizations have shared with us how their needs have evolved from not only wanting to secure existing applications, but wanting to build new ones. AI has allowed non-technical teams to design, build, and scale tools tailored specifically for their workstreams and to advance their mission.
This opportunity is arriving at a challenging moment for the sector. Many organizations report operating under financial strain after major reductions in government funding contributed to layoffs and closing of programs. As groups rebuild their work in a new environment, some have reported using AI to help do more with less. But adoption remains ad hoc. In a CIVICUS survey, more than half of civil society respondents viewed privacy concerns as a barrier to use. These groups hold sensitive data, like the location of activists or the identities of anonymous sources, and are disproportionately at risk of cyberattacks, according to our own Cloudflare data. Additionally, the cost of building and running complex automation workflows can be prohibitive. Among those surveyed by CIVICUS, 48% cited financial constraints limiting their adoption.
Cloudflare helps address these concerns. Our developer services incorporate security and privacy protections from the outset. By running on our global network, applications are automatically protected from distributed denial-of-service attacks and attempts to gain unauthorized access to internal systems. Civil society groups can also build and run applications more cost effectively. Our lightweight and serverless architecture scales with demand without requiring organizations to provision or pay for idle infrastructure. Workers AI also removes the need to operate dedicated GPU infrastructure, while AI Gateway provides rate limits, observability into how teams are using AI, and intelligent model routing to reduce unnecessary model calls and keep costs under control.
During Birthday Week last year, we expanded Cloudflare for Startups to include non-profit and public interest organizations, providing each of them with up to $250,000 in credits for our developer services. We are proud to announce that we selected 30 organizations to participate in our first non-profit, startup cohort, and we have been thrilled to watch their ideas and tools designed to tackle problems in climate science, humanitarian aid, mental health, civic engagement, and education come to life.
Here is what a few of these non-profits have built.
We also heard from larger civil society organizations who wanted to build complex workflows and welcomed extra engineering support. These projects included several variations of the same problem: large amounts of manually-collected, dispersed information that needed to be integrated, reviewed, and structured in order for effective analysis to be conducted. The data involved was also sensitive, like the names of human rights abuse victims, and the risk of unauthorized access, data leakage, or hallucinations in an output could have serious consequences.
Cloudflare not only provided free access to its developer service to support the development of each of these tools, but also assembled a team of volunteer engineers — product experts, front-end designers, back-end builders, and security advisors — to design and build them. Questions we’re exploring include where automation is helpful, which models align best with each task, how to incorporate the necessary security controls, and where a person must remain involved.
Prototype of Freedom House’s tool tracking transnational repression
Prototype of GNI’s dashboard that tracks the status of engagement opportunities
It’s incredible to see how civil society organizations are evolving their work in an era of AI. Cloudflare is excited to play a small role in this process. Working directly alongside these organizations not only helps advance their missions, but also helps inform how we think about the security and privacy in high-risk environments and how we can scale similar programs moving forward.
Our first non-profit startup cohort shows how automation can help the next generation of community service organizations use AI and automation to serve the public, and how they can do this securely and affordably.
We’re excited to announce that starting today we are officially opening our startup program to our second cohort of non-profit organizations.
If your organization is interested, apply here and select the non-profit checkbox. We’d love to build with you!
Continue reading on the original blog to support the author
Read full articleThis update provides critical visibility and governance for AI workloads. By linking requests to identities and baselining behavior, engineers can prevent runaway costs from rogue agents and enforce security policies without building custom authentication for every AI client.
Cloudflare Wallets solve the friction of AI agents accessing paid services. By providing machine-native payments via the x402 protocol and programmable guardrails, engineers can build autonomous agents that safely discover, test, and purchase APIs without manual human intervention.
Uncontrolled AI spend is a major challenge for organizations. These tools provide the observability and governance needed to scale AI usage sustainably by offering granular cost attribution and automated guardrails to prevent unexpected bill shock.
This incident highlights a critical security risk in LLM inference: stale KV cache data leaking across tenants. It demonstrates how subtle bugs in disaggregated architectures can bypass isolation, emphasizing the need for strict memory management and configuration validation.
