Develop Your Organisational AI Readiness
Develop Your Organisational AI Readiness: A Practical Guide for Nonprofits
Why This Guide Matters
Many nonprofit organisations are eager to explore artificial intelligence but do not know where to start. The Humanitarian Leadership Academy and NetHope developed this quick-start guide to address exactly that gap. It builds on a January 2026 webinar, "Humanitarian AI: Lessons learned, trends and opportunities for 2026," which brought together experts and practitioners to share experiences and practical advice.
The core message:
Start with ethical guardrails, build capacity across your entire organisation, test and learn, then scale what works. Do not let perfect be the enemy of good — but do not skip the foundational steps either.
Nine Practical Steps to Build Organisational AI Readiness
The guide outlines a clear, sequential approach that any nonprofit can adapt to its own context, size, and resources. Each step builds on the previous one, but organisations can enter at different points depending on their maturity.
Step 1: Agree your responsible AI framework
Set clear guardrails on what your organisation will and will not use AI for. NetHope's Humanitarian AI Code of Conduct provides a solid starting point. Create an AI policy that enables staff to try out AI tools while being very clear about permitted and prohibited uses. This policy must protect both your data and your organisation's reputation. Avoid the spread of unofficial "shadow AI" tools by providing a list of recommended, safe tools, creating a process to assess user-recommended tools, and discouraging the use of unapproved tools.
Step 2: Build capacity everywhere in your organisation
Generative AI tools are widely accessible, and AI functionality is increasingly built into the software packages you already use. Ensure that everyone in your organisation knows not just how to use the tools, but also their limitations and risks, and how to mitigate those risks. Many free learning resources are available, including NetHope's "Unlocking AI for NGOs" course hosted on the HLA's Kaya learning platform, available in both English and Spanish.
Step 3: Test, learn and share
Encourage users who have completed foundational AI training to experiment with AI tools in their day-to-day work. Invite them to capture outcomes and learnings, and to share their use cases and prompts with others in the organisation and beyond. Establish a support mechanism — whether a monthly AI hackathon, a dedicated chat channel, or a knowledge sharing site.
Step 4: Define your AI purpose
Have a cross-departmental conversation about what your best use cases for AI might be and how you might prioritise and scale them. Is your key objective to achieve more operational efficiency? Or could AI transform the way you provide humanitarian services? Use tools such as NetHope's AI Suitability Toolkit to guide your discussions.
Step 5: Start small but think big
The real benefits for your organisation will only come if you manage to scale good use cases and rethink your internal processes and external service delivery — truly transforming how you work with AI. Once you have built basic organisational AI capacity, start thinking about what it takes to scale.
Step 6: Embed and measure
Everyone in the sector understands behaviour change and impact measurement — apply those same concepts to your AI projects. Employ change management tools for successful adoption at scale. Measure the outcomes of the AI tools you implement and compare them with the total cost of ownership. Stop projects that do not deliver value.
Step 7: Look for partners
Building the in-house capabilities and capacity to implement AI solutions that go beyond standard, built-in AI chatbots and functionality is hard and costly. Identify partners who understand your needs and can truly support your journey. These might include technologists, consultants, universities, or other nonprofit organisations.
Step 8: Don't lose focus on your data
If you want to implement AI applications, you must have good data governance and data management practices in place. If you do not, now is the time to change that. Use NetHope's Data Governance Toolkit for Nonprofits to set those data foundations — they are critical to making your AI project work. Before getting started with agentic AI, ask critical questions: What processes would you want to automate? Are your underlying data, systems and infrastructure ready to use AI agents responsibly and safely?
Step 9: Participate in sector collaboration on AI
Open conversations about what worked and what did not are the most helpful learning opportunities for humanitarian staff and lift the overall capability of the sector. Share your approaches and use cases with each other. Collaborate to create common solutions that will serve all humanitarians. Join forces to advocate for AI offerings that close, not widen, the digital divide. Connect through webinars such as the HLA's Humanitarian Tech Webinar Series, and form communities of practice. Nonprofits of all sizes can join NetHope's community. NetHope Members can join the AI and Emerging Technologies Working Group. NetHope Connected enables smaller organisations to develop their AI approaches together. Take the NetHope Digital Nonprofit AI Readiness Assessment to see how you score compared to the sector benchmark.
Key Resources for Nonprofits
NetHope Humanitarian AI Code of Conduct
A starting point for organisations to establish their own ethical guardrails and responsible AI framework.
Unlocking AI for NGOs (free course)
Hosted on HLA's Kaya learning platform, available in English and Spanish. Provides foundational AI skills for nonprofit staff.
NetHope AI Suitability Toolkit
Helps organisations identify and prioritise the most promising AI use cases for their specific context.
Data Governance Toolkit for Nonprofits
Essential resource for establishing the data foundations required for successful AI implementation.
NetHope AI Lighthouse
A comprehensive guide to responsible, purpose-driven AI specifically for nonprofits.
Digital Nonprofit AI Readiness Assessment
A benchmarking tool to see how your organisation scores on AI readiness compared to the sector.
Key Takeaways for M&E and Capacity Strengthening Professionals
Shadow AI is a real risk
The guide explicitly warns against unofficial AI use. Evaluators should assess whether organisations have policies to prevent staff from using unapproved tools that could expose sensitive data.
Capacity building must be organisation-wide
AI literacy is not just for technical staff. Everyone needs to understand both the capabilities and the limitations of AI tools, as well as how to mitigate risks.
Start small, measure, and scale or stop
Apply standard M&E principles to AI projects: define outcomes, measure impact, compare with costs, and stop initiatives that do not deliver value.
Data governance is a prerequisite
The guide is clear: without good data governance and management practices, AI projects will fail. This is a critical readiness assessment question for any organisation.
Sector collaboration as a force multiplier
The guide emphasises that open conversations about successes and failures are the most helpful learning opportunities for humanitarian staff. Sharing approaches, collaborating on common solutions, and advocating together for AI offerings that close rather than widen the digital divide can lift the entire sector's capability. Nonprofits of all sizes are encouraged to join NetHope's community, participate in working groups, and use shared resources rather than starting from scratch.
Frequently Asked Questions
What is "shadow AI" and why should we care?
Shadow AI refers to the use of unauthorised AI tools by staff without organisational approval or oversight. The guide warns that this practice can expose sensitive data and create governance risks. The solution is to provide a list of approved, safe tools and a process for assessing new tools that users recommend, while discouraging the use of unapproved tools.
Does every nonprofit need a custom AI policy?
Yes, but you do not need to start from scratch. NetHope's Humanitarian AI Code of Conduct provides a solid foundation that you can adapt to your organisation's specific context, risk tolerance, and data sensitivity requirements. The key is to be very clear about what AI can and cannot be used for.
What free training resources are available?
NetHope's "Unlocking AI for NGOs" course is available for free on the HLA's Kaya learning platform in both English and Spanish. The HLA also offers webinars, podcasts, and research products through its Resources Hub. Kaya also offers HPass digital badges to recognise learning achievements.
How do we know if we are ready for AI?
The guide recommends taking the NetHope Digital Nonprofit AI Readiness Assessment to benchmark your organisation against the sector. Key readiness indicators include having a responsible AI framework, basic AI literacy across staff, good data governance practices, and a clear understanding of your AI purpose and priority use cases.
Starting your AI journey: Practical, not perfect
The HLA and NetHope guide offers a clear, actionable framework for nonprofits at any stage of AI readiness. The message is consistent: start with ethics and guardrails, build capacity across the whole organisation, test and learn, then scale what works.
You do not need to have all the answers before you begin. But you do need to be intentional about the risks, transparent about your limitations, and committed to measuring what works.
As the guide notes, the real benefits come when organisations manage to scale good use cases and rethink their internal processes and external service delivery — truly transforming how they work with AI.
Related Resources
