Week of 21/10/2024 Advancing Humanitarian Aid with AI: Exploring Innovative Tools, Ethical Challenges, and Community Engagement

Artificial intelligence (AI) is reshaping the humanitarian aid and development sectors, offering powerful new tools while raising critical ethical challenges. This week, we explore AI-driven platforms like GANNET that enhance crisis response, the pressing need to safeguard children online, and the role of MERL Tech in promoting ethical, community-focused AI development. Plus, we spotlight the soft launch of AidGPT.org and invite you to join our upcoming AI in Humanitarian Informal Hangout.

Exploring GANNET: An AI-Powered Tool for Humanitarian Response

This week, I had the opportunity to explore GANNET, an AI-powered platform developed by Data Friendly Space (DFS) specifically for humanitarian workers and organizations. I believe this is the beginning of exciting use cases for AI in humanitarian responses.

Key Features of GANNET:

• Real-Time Data and Analytics: Provides up-to-date information from reliable sources to help humanitarians make quick decisions in crisis situations.
• All-in-One Platform: Combines AI, analytics, visualization, monitoring, and alerts in a single interface, streamlining workflows.
• Data Exploration: Allows users to delve into humanitarian data through interactive maps, charts, and filters. The platform contains over 112,000 documents from 165 sources across six categories.
• AI-Powered Querying: Enables natural language questions with AI-generated responses based on available data. For example, you can inquire about key events in the Sudan crisis and receive a detailed summary.
• Collaboration Tools: Facilitates the creation of "threads" for teams to collaborate and share insights seamlessly.
• Trust and Reliability: Emphasizes providing trustworthy and actionable data to support informed decision-making.

I found GANNET to be a exciting tool that could significantly enhance how we access, analyze, and collaborate on critical information during humanitarian crises. Its real-time updates and AI capabilities make it a promising resource for improving response times and effectiveness.

Call to Action: I highly recommend humanitarian professionals and organizations check out GANNET to explore its potential benefits firsthand. Visit gannet.ai to learn more and consider integrating this innovative tool into your humanitarian response efforts.

By embracing platforms like GANNET, we can leverage technology to make our work more efficient and impactful, ultimately better serving the communities in need.

Childproofing AI: A Critical Blindspot in the Social Sector

This week, I want to highlight a crucial issue raised by Isabelle Amazon-Brown in her MERL Tech blog post, "Childproofing AI: a collective blindspot for the social sector?." Despite children constituting one-third of internet users globally and being avid adopters of new technologies, the impact of AI on young users is often overlooked. Children face unique risks from AI tools like chatbots, including exposure to harmful content, privacy violations, and potential negative effects on their cognitive and emotional development.

Amazon-Brown emphasizes that AI systems often fail to interpret children's non-literal, playful language, which can hinder critical thinking and creativity. The lack of specific legal frameworks exacerbates these risks, making it imperative for our sector to prioritize child-centric design and advocate for stronger protections.

Call to Action: I strongly encourage you to read Isabelle's insightful blog post here to understand the depth of this issue. Additionally, explore the detailed papers by Dr. Nomisha Kurian and Boglárka Jánoskúti for practical recommendations on safeguarding children in AI environments. Let's collaborate to ensure that AI technologies are safe and beneficial for our youngest users.

• Dr. Nomisha Kurian's Paper: “No, Alexa, no!”: Designing Child-Safe AI and Protecting Children from the Risks of the “Empathy Gap” in Large Language Models
• Safeguarding Guidance: Practical Guidance for Safeguarding Mechanisms in AI and Non-AI Chatbots

By engaging with these resources, we can work towards childproofing AI and fostering a safer digital environment for children worldwide.

Spotlight on MERL Tech: Advancing Technology in Monitoring and Evaluation

This week, I want to highlight the excellent work of MERL Tech, an initiative dedicated to enhancing Monitoring, Evaluation, Research, and Learning (MERL) through digital tools and approaches. MERL Tech collaborates with social sector organizations to promote thoughtful, tech-enabled program design and implementation. They convene communities of practice to share knowledge and advance responsible, ethical digital methods that prioritize people and communities.

Past Work and Resources:

MERL Tech has significantly contributed to the field with their "State of the Field" series, which includes insightful papers such as:

• The Evolution of MERL Tech: Exploring how technology has transformed MERL practices over time.
• Big Data to Data Science: Discussing the shift from traditional data collection to advanced data analytics in the MERL space.
• Emerging Technologies in MERL: Highlighting new tools and methodologies for international development programs.

They also offer valuable resources on data ethics, responsible data governance, and indigenous data rights, providing guidance to practitioners navigating the complexities of digital data use.

Upcoming Events:

MERL Tech has several exciting events on the horizon:

• Tech and Eval Mixer at the AEA Conference: On October 23, join fellow professionals at the Mercy Corps Offices for a networking mixer following the American Evaluation Association's poster session in Portland, OR.
• AI Sessions at the AEA Conference: From October 21-26, MERL Tech and the NLP Community of Practice are organizing multiple sessions on artificial intelligence at the AEA conference.
• EES Webinar on AI and Evaluation: On November 4, the European Evaluation Society invites you to a free webinar exploring the intersection of AI and evaluation from 2-4 pm CET.
• Humanitarian AI + MERL Working Group Meeting: On November 7, the NLP Community of Practice is launching a new working group focused on AI's role in humanitarian monitoring and evaluation from 10:00 am to 11:30 am ET.

Get Involved:

I highly recommend checking out MERL Tech's work and participating in their upcoming events. Their efforts are instrumental in advancing the effective and ethical use of technology in our field. You can learn more and stay updated by visiting their website at merltech.org and subscribing to their newsletter.

CALP Takes Intentional Steps with AI

The Cash Learning Partnership (CALP) has taken a proactive approach to integrating Artificial Intelligence (AI) into its operations. Recognizing the rapid advancement of AI technologies, CALP formed an internal task team and partnered with Torchbox to conduct a comprehensive staff consultation process. This initiative resulted in two key outcomes:

1. AI Principles: CALP developed six guiding principles to balance the benefits and risks of AI implementation. These principles emphasize human oversight, ethical considerations, data protection, transparency, efficiency, and maintaining human expertise.
2. AI Roadmap: Rather than a strict deployment plan, CALP created a flexible "atlas" of potential AI use cases across different teams. This approach allows for exploration and prioritization of tools that best suit specific needs, acknowledging current funding limitations for expensive AI licenses.

By formalizing their approach to AI, CALP demonstrates a commitment to ethical and appropriate integration of these technologies in their work. This strategy empowers teams to innovate while adhering to carefully considered principles, ensuring that AI adoption aligns with CALP's mission and values.

Introducing AidGPT.org: A New Forum for AI in Humanitarian Aid and Development

Following a request at our last AI and Humanitarian informal hangout, I'm excited to announce that my team and I at MarketImpact have launched AidGPT.org—a new online forum dedicated to exploring how artificial intelligence can revolutionize humanitarian assistance and development work.

What is AidGPT.org?

AidGPT.org is a unique platform serving as a central hub for humanitarian professionals, technologists, researchers, and enthusiasts passionate about leveraging innovation to tackle complex global challenges. It's an open, non-commercial space designed to foster collaboration and share insights on AI applications in the humanitarian and development sectors.

Our Vision

• Foster Collaboration: Bring together diverse experts to create innovative solutions for humanitarian challenges.
• Promote Ethical AI: Address ethical considerations in AI use within crisis settings and humanitarian operations.
• Open Discussion: Provide a space for sharing ideas, challenges, and lessons learned.

Why Join AidGPT.org?

• Connect with Peers: Engage with a community of like-minded professionals and enthusiasts.
• Stay Informed: Keep up-to-date with the latest developments, projects, and best practices in AI for humanitarian aid.
• Contribute: Share your insights, experiences, and questions to help advance the field.

Join Our Community

Be part of the movement to harness the power of AI for humanitarian causes. Join our forum to collaborate, learn, and make a difference.

👉 Join AidGPT.org Forum

Join Our Next AI in Humanitarian & Development Discussion – October 23rd

Following our lively and insightful first discussion on September 12th, I'm excited to invite you to our next informal AI discussion in the humanitarian and development space.

🗓 Date: Wednesday, October 23rd ⏰ Time: 4:00 PM (Amman Time) 📍 Sign Up Here: Event Registration

During our last chat, we covered a wide range of topics, including ethical AI considerations, practical use cases, data protection challenges, and the opportunities and risks of AI in humanitarian work. It was fantastic to hear from such a diverse group, sharing insights and experiences across different sectors and contexts.

For this follow-up discussion, we aim to cover:

• Recap of Key Takeaways: Reflecting on insights from our last session.
• Deep Dive into Practical AI Use Cases: Sharing lessons learned and best practices.
• Addressing Emerging Concerns: Focusing on AI data governance and consent issues.
• Open Q&A Session: An opportunity to share updates, challenges, and success stories from your AI projects.

As always, this will be an open space under Chatham House rules—feel free to speak openly, but please don't attribute comments to specific individuals outside this meeting.

Call to Action: I encourage everyone interested in the intersection of AI and humanitarian/development work to join us for this engaging session. Feel free to invite others who might be interested. If you have specific questions or topics you'd like to address, please share them with me beforehand.

Speaking at the AI for Development Conference in Barcelona – October 28, 2024

And finally I'm thrilled to announce that I'll be speaking at the upcoming AI for Development Conference in Barcelona on October 28, 2024! This event will gather global thought leaders to explore how artificial intelligence can drive positive impact in crisis response and sustainable development.

During my breakout session, I'll be sharing insights on "Artificial Intelligence in Crisis Response: Improving Cash Assistance Efficiency and Accountability." Building on my recent work with NRC/CashCap as their MENA regional expert, I'll delve into how AI can revolutionize humanitarian programs—particularly in enhancing Multi-Purpose Cash Assistance (MPCA) delivery—while addressing key ethical considerations like data privacy, algorithmic bias, and digital inclusion.

I'm excited to join this vibrant community of innovators and experts to discuss the future of AI in the humanitarian sector. It's an excellent opportunity to engage with innovators and experts passionate about leveraging technology for good.

Call to Action: If you're interested in how AI can transform humanitarian aid and want to participate in meaningful discussions, I encourage you to join us at the conference.

👉 Register now: AI4D Conference Registration

Looking forward to connecting with many of you there!