Humanitarian AI Research Fellowship 2026 at NetHope: Remote, Part-Time Fellowship for AI Researchers and Humanitarian Practitioners
Artificial intelligence is increasingly being explored by humanitarian, development and nonprofit organizations for tasks ranging from data analysis and program support to communications, operational decision-making and service delivery. At the same time, organizations working with vulnerable communities face particular challenges around data protection, responsible AI, governance, connectivity, localization and the practical limitations of AI systems.
To address some of these issues, NetHope has launched a new Humanitarian AI Research Fellowship designed for researchers and experienced practitioners interested in examining how artificial intelligence can be used responsibly in humanitarian and development settings.
The fellowship is part-time, fully remote and internationally open, and will bring together a small cohort of approximately four to six fellows. Participants will conduct independent research within one of two research streams, with guidance from the AI Fellowship Lead and periodic engagement with NetHope subject-matter experts.
Applications close on October 9, 2026, although NetHope has indicated that it may review applications and select fellows before the deadline. Applicants are therefore encouraged to submit early.
About the NetHope Humanitarian AI Research Fellowship
The Humanitarian AI Research Fellowship is a short-term research initiative intended to expand the number of researchers and practitioners who understand the opportunities, limitations and risks associated with artificial intelligence in humanitarian and development work.
Rather than focusing solely on theoretical AI research, the fellowship is intended to generate findings that humanitarian organizations, nonprofit organizations and other practitioners can use in real-world settings.
NetHope says the research will focus on practical questions around AI adoption, governance, safety, localization and technical implementation. The fellowship therefore sits at the intersection of artificial intelligence, humanitarian action, nonprofit technology, responsible AI and international development.
NetHope brings together humanitarian, development and conservation organizations around technology-related challenges. Its network includes more than 60 member organizations working across more than 190 countries, alongside collaboration with more than 70 technology companies.
The new fellowship builds on NetHope’s wider work around responsible technology and AI. The organization has been developing AI-related toolkits, research and initiatives aimed at helping nonprofits understand and responsibly implement emerging technologies.
Key Details at a Glance
- Organization: NetHope
- Opportunity: Humanitarian AI Research Fellowship
- Format: Fully remote
- Work arrangement: Part-time
- Time commitment: Approximately 10 hours per week
- Fellowship period: Mid-November 2026 through February 2027
- Expected cohort: Approximately 4–6 fellows
- Location: Internationally open
- Stipend: Provided to fellows; the exact amount is not publicly listed
- Research areas: AI Adoption & Governance; AI Systems & Localization
- Authorship: Fellows receive named authorship on published research
- Application deadline: October 9, 2026
- Early selection: NetHope may select fellows before the application deadline
What Will Humanitarian AI Research Fellows Work On?
The fellowship has two parallel research streams. Applicants should have a background that aligns with one of these areas.
Stream 1: AI Adoption and Governance
The first research stream focuses on understanding how humanitarian and nonprofit organizations are actually adopting artificial intelligence, how they govern its use and how they manage the risks associated with AI.
The work is intended to go beyond general discussions about responsible AI by examining what happens inside organizations when AI tools are introduced into real operational and program environments.
Possible areas of research include:
- Documenting real-world experiences of humanitarian and nonprofit organizations integrating AI into their operations and programs.
- Examining how organizations assess their readiness for AI adoption.
- Understanding how AI adoption occurs at the level of individual implementing partners.
- Comparing different organizational approaches to AI adoption.
- Examining direct partnerships with AI vendors.
- Studying open-source and internally developed AI systems.
- Examining the use of commercially available, off-the-shelf AI tools.
- Identifying the factors that influence an organization’s choice of AI approach.
- Identifying major AI safety and governance risks within humanitarian organizations.
- Exploring approaches for identifying and mitigating AI-related risks.
- Examining incident-response practices associated with AI systems.
- Developing case studies that capture useful or cautionary lessons from organizations using AI.
Fellows working in this stream are expected to develop at least two case studies examining how humanitarian or nonprofit organizations are adopting, governing and managing risks around AI tools.
Who fits Stream 1?
This stream is particularly relevant to people with backgrounds in:
- Policy
- Program evaluation
- Humanitarian systems
- International development
- Nonprofit operations
- Organizational change
- Technology adoption
- AI governance
- Responsible technology
Experience with technology adoption or organizational change is considered an advantage.
Stream 2: AI Systems and Localization
The second stream takes a more technical approach.
It focuses on AI systems that can operate effectively in humanitarian environments where factors such as limited connectivity, local language requirements, resource constraints and offline access may affect how technology can be deployed.
This is particularly relevant because humanitarian operations do not always take place in environments with reliable high-speed internet, powerful computing infrastructure or access to the latest proprietary AI platforms.
Potential research activities include:
- Evaluating open-source AI models for humanitarian applications.
- Evaluating open-weight AI models.
- Assessing localized AI models.
- Examining fine-tuned AI models for humanitarian use cases.
- Comparing open or locally deployed AI systems with closed proprietary systems.
- Studying the risks associated with different AI deployment approaches.
- Examining the advantages and limitations of localized AI systems.
- Identifying practical AI applications for resource-constrained environments.
- Identifying gaps in organizational readiness for AI deployment.
- Developing a small collection of simple, user-friendly AI tools.
- Contributing to libraries of supervised fine-tuning prompts.
- Contributing datasets for humanitarian AI applications.
This stream is therefore particularly suited to applicants with technical expertise who are interested in applying AI to real-world humanitarian problems rather than focusing exclusively on laboratory or commercial applications.
Who fits Stream 2?
Applicants are expected to have backgrounds in areas such as:
- Computer science
- Machine learning
- Applied data science
- Artificial intelligence
- AI evaluation
- Data analysis
- Model testing
- AI safety
Knowledge of or interest in the humanitarian sector is relevant, while experience with model evaluation or red teaming is an additional advantage.
Research Standards and Responsible AI Requirements
Because the fellowship involves research connected to humanitarian organizations and potentially affected communities, fellows will be expected to follow established research, data protection and AI governance requirements.
For example, case studies involving NetHope members or partners require appropriate consent. NetHope will also review findings before publication.
Fellows will be required to follow NetHope’s AI Policy, including completing an AI Disclosure Form before beginning the fellowship.
They must also follow NetHope’s Data Protection Policy.
Where research involves personal information or affected communities, the research must comply with applicable data protection and research ethics requirements.
These requirements mean that applicants should be comfortable handling research information responsibly and understanding that humanitarian AI research can involve sensitive operational and community-related information.
Fellowship Duration and Workload
The fellowship is designed as a part-time commitment rather than a full-time position.
Selected fellows are expected to contribute approximately 10 hours per week.
The fellowship is scheduled to run from approximately mid-November 2026 through February 2027.
This structure allows researchers and experienced practitioners to participate while continuing other professional, academic or research activities, subject to their ability to meet the fellowship’s research commitments.
The expected cohort size is relatively small, with approximately 4 to 6 fellows selected.
What Support Will Fellows Receive?
Fellows will conduct independent research within their selected stream, but they will not be completely isolated.
The fellowship provides guidance from the AI Fellowship Lead, together with periodic check-ins and engagement with NetHope subject-matter experts.
This structure is intended to provide researchers with access to sector knowledge while allowing them to take responsibility for their own research projects.
The fellowship also provides an opportunity to work on research that can potentially be used by NetHope member organizations and the wider nonprofit community.
Stipend and Research Authorship
The fellowship provides a stipend in recognition of fellows’ time and contributions.
However, NetHope has not publicly stated the exact stipend amount. The available fellowship information says that stipend details will be provided to selected fellows and included in the fellowship agreement.
Another significant feature is research authorship.
Fellows will receive named authorship credit on published research arising from the fellowship.
This can provide researchers and practitioners with an opportunity to contribute to published work related to responsible AI, humanitarian technology and nonprofit innovation.
Applicants should also understand that NetHope will own the research outputs and associated intellectual property, including relevant datasets, prompts, tools and code. Release and licensing arrangements will be specified in the fellowship agreement.
Who Should Apply?
NetHope says it is looking for mid-career to seasoned practitioners who have strong writing and analytical abilities and a genuine interest in the role of responsible AI in humanitarian impact.
The opportunity is therefore not presented primarily as an entry-level AI internship.
Applicants should be able to demonstrate that they can independently investigate a research question, analyze information, communicate findings clearly and complete a research project within a defined timeframe.
Across both research streams, applicants should demonstrate:
- Strong written communication.
- Strong analytical skills.
- Ability to work independently.
- Ability to manage an independent research project.
- Ability to work within a fixed research timeline.
- Genuine interest in responsible AI.
- Interest in humanitarian, nonprofit or development applications of technology.
Application Requirements
According to the published fellowship information, applicants should prepare several components for their application.
1. Curriculum Vitae
Applicants should submit a CV that clearly demonstrates their relevant professional, research, analytical, technical or humanitarian experience.
The CV should make it easy for reviewers to identify experience related to the selected research stream.
2. Letter of Interest
The application requires a letter of interest addressing several questions.
Applicants are asked to explain:
Motivation and relevant experience
Applicants should describe their motivation for participating in the fellowship, relevant experience and interest in humanitarian and nonprofit work.
AI risk in the humanitarian sector
Applicants should identify one significant concern or risk associated with AI in humanitarian work and briefly explain a possible mitigation approach.
Real-world AI experience
Applicants should discuss an AI tool they have actually used or evaluated in a real project, rather than simply demonstrating or experimenting with it.
They should explain one way the AI tool underperformed in the actual context where it was used and how they identified the problem.
Applicants are specifically advised not to disclose confidential information belonging to employers or clients.
3. Professional References
Applicants should provide two to three professional references.
These should ideally be people who can speak credibly about the applicant’s professional abilities, research skills, analytical work, technical experience or ability to work independently.
4. Optional Writing Sample
A writing sample may also be submitted.
A strong writing sample could help demonstrate the applicant’s ability to analyze a subject, conduct research and communicate findings clearly.
Important Selection Requirement: OFAC Screening
All selected fellows must pass an OFAC screening before starting the fellowship.
This is a condition of participation and should be taken into consideration by applicants from all locations.
The fellowship is described as internationally open, but international openness does not remove the requirement to satisfy NetHope’s screening and compliance requirements.
Is the Fellowship Open Internationally?
Yes.
The opportunity is explicitly described as internationally open and remote, meaning applicants are not restricted to a particular country or continent based on the published eligibility information.
Applicants can therefore apply from different parts of the world, provided they meet the professional requirements and the fellowship’s screening and compliance conditions.
The announcement does not specify a particular race, ethnicity or nationality as an eligibility requirement. It is also not presented as an opportunity restricted to applicants from Africa, Asia, Europe, North America or another particular continent.
However, all selected fellows must pass the required OFAC screening before beginning the fellowship.
What Makes an Application Relevant to the Fellowship?
Applicants should make a clear connection between their experience and one of the two research streams.
For example, someone applying to AI Adoption & Governance could emphasize experience involving:
- Policy research
- Program evaluation
- Technology adoption
- Organizational change
- Humanitarian programs
- International development
- AI governance
- Responsible technology
- Research and report writing
- Data protection
- Risk assessment
Someone applying to AI Systems & Localization could emphasize:
- Machine learning
- Artificial intelligence
- Data science
- Model evaluation
- AI testing
- Dataset development
- Prompt engineering
- Fine-tuning
- AI safety
- Red teaming
- Open-source AI
- Localized AI applications
- Technical research
The strongest applications are likely to demonstrate not simply an interest in AI, but an ability to analyze how AI performs in real environments and communicate practical findings.
Why the Humanitarian Context Matters
AI systems that work well in well-connected commercial environments may face different challenges when deployed in humanitarian settings.
Humanitarian organizations may operate in environments affected by:
- Limited internet connectivity
- Limited computing resources
- Data protection concerns
- Sensitive personal information
- Multiple languages
- Local cultural contexts
- Emergency situations
- Resource constraints
- Unreliable infrastructure
- Requirements for responsible use of technology
NetHope’s own research has highlighted gaps in AI governance relevant to nonprofit organizations, including issues around humanitarian principles, data protection in low-infrastructure settings and relationships between funders and organizations receiving funding.
The fellowship therefore creates an opportunity to examine these issues through practical research rather than treating AI governance as a purely theoretical subject.
Important Dates
Application Deadline
October 9, 2026
Applicants should be aware that NetHope may review applications and select fellows before the stated closing date because of the anticipated volume of applications.
Fellowship Period
Mid-November 2026 through February 2027
Weekly Commitment
Approximately 10 hours per week
Expected Cohort
4–6 fellows
How to Apply
Applicants should prepare their CV, letter of interest, professional references and, if available, a relevant writing sample.
The application should clearly identify the applicant’s motivation, relevant experience, understanding of responsible AI and experience using or evaluating an AI tool in a real project.
The official announcement and application links should be checked before submitting, particularly because NetHope may make selections before the October 9 deadline.
For Official information visit NetHope
Conclusion
The NetHope Humanitarian AI Research Fellowship 2026 is a short-term, part-time research opportunity focused on the practical use of artificial intelligence in humanitarian and nonprofit settings.
The fellowship offers two distinct directions. The AI Adoption & Governance stream examines how organizations introduce AI, manage organizational change and address associated safety and governance risks. The AI Systems & Localization stream focuses more heavily on evaluating AI systems and developing practical approaches for localized, offline and low-connectivity environments.
The fellowship is scheduled for approximately mid-November 2026 through February 2027 and requires around 10 hours of work per week. Selected fellows will receive a stipend, guidance from NetHope’s AI Fellowship Lead and subject-matter experts, and named authorship on published research.
The opportunity is remote and internationally open, with no specific race, nationality or continent identified as an eligibility restriction in the published requirements. However, applicants must meet the relevant professional requirements for their chosen stream and selected fellows must pass OFAC screening before beginning.
With applications closing on October 9, 2026, prospective applicants should prepare their materials early, particularly because NetHope may make selections before the official deadline.
For more global fellowship program visit OFY: https://opportunitiesforyouth.org/category/fellowship/
