Last updated 2 months ago
Misinformation spreads fast in local languages, causing fear and conflict. There’s no real-time, multilingual tool to detect and verify false content at scale.
PeaceGuard AI is a mobile and web app that detects, verifies, and proactively alerts users about false content in real time, supporting local languages and low-bandwidth use.
This is the total amount allocated to PEACEGUARD AI - A MULTILINGUAL MISINFORMATION DETECTION & VE.
Please provide your proposal title
PEACEGUARD AI - A MULTILINGUAL MISINFORMATION DETECTION & VE
Enter the amount of funding you are requesting in ADA
88500
Please specify how many months you expect your project to last
9
Please indicate if your proposal has been auto-translated
No
Original Language
en
What is the problem you want to solve?
Misinformation spreads fast in local languages, causing fear and conflict. There’s no real-time, multilingual tool to detect and verify false content at scale.
Supporting links
Does your project have any dependencies on other organizations, technical or otherwise?
No
Describe any dependencies or write 'No dependencies'
No critical dependencies. The project will be executed independently by the core team. While we may collaborate with fact-checkers or local partners, the MVP build and delivery do not depend on external approvals or other proposals.
Will your project's outputs be fully open source?
Yes
License and Additional Information
This project will be fully open-source to ensure transparency, community collaboration, and long-term sustainability. All documentation will be made publicly available on GitHub, enabling developers, researchers, and community members to review, contribute, and adapt the solution for their specific needs. The project will adopt the MIT License, which allows anyone to use, modify, and distribute the software with minimal restrictions while maintaining attribution to the original creators.
Please choose the most relevant theme and tag related to the outcomes of your proposal.
AI
Describe what makes your idea innovative compared to what has been previously funded (whether by you or others).
PeaceGuard AI is the first multilingual, real-time misinformation detection tool. Unlike previous projects, it verifies not just text, but also video and speech, and includes proactive alerts to stop misinformation before it spreads. It’s mobile-friendly, works in low-bandwidth environments, and empowers grassroots users, bridging the gap between AI, peacebuilding, and accessibility.
Describe what your prototype or MVP will demonstrate, and where it can be accessed.
The MVP will demonstrate real-time detection and verification of false content across text, speech, and video in at least two African languages. Users will be able to input content and receive instant truth analysis or alerts. The MVP will be accessible via a mobile app (Android-first) and a lightweight web platform. It will include a user dashboard, local language support, and basic reporting features. Early access will be provided to pilot partners, with a public demo available online.
Describe realistic measures of success, ideally with on-chain metrics.
Success will be measured by:
Launch of MVP with support for at least 2 African languages
at least 50 verified users interacting with the tool
Detection of 50+ misinformation cases during pilot
3+ partnerships with fact-checking or peacebuilding organizations
User feedback showing 80%+ satisfaction
Public demo and report published
Onchain metrics (future phases):
Adoption of ADA-based micro-payments for API access
Integration of decentralized identity (DID) for verified users
Potential issuance of trust scores or truth tokens for verified facts
Please describe your proposed solution and how it addresses the problem
PeaceGuard AI is a multilingual mobile and web-based application that uses advanced AI to detect and verify misinformation in real-time. The platform supports the following core features:
Real-time detection of harmful or misleading content.
Fact-checking of written and spoken statements.
Verification of video authenticity, using metadata and source analysis.
Proactive alerts on trending disinformation before it spreads widely.
PeaceGuard AI starts with Africa, where the need is urgent, and the opportunity to save lives is real. But it’s built for everyone.
From day one, anyone worldwide can access basic features. As the platform grows, language and region support will expand, with Africa as the innovation hub for truth-tech.
Please define the positive impact your project will have on the wider Cardano community
PeaceGuard AI will create significant positive impact on the Cardano ecosystem in multiple, interconnected ways, across adoption, innovation, infrastructure, and narrative. This project doesn't just build a tool; it opens new doors for Cardano to lead in responsible, real-world AI, social impact, and decentralized digital trust.
What is your capability to deliver your project with high levels of trust and accountability? How do you intend to validate if your approach is feasible?
PeaceGuard AI will be led by Ayomide Osemene, a digital educator, virtual assistant trainer, and Web3 advocate with hands-on experience delivering high-impact community projects across Africa. Ayomide has participated in the AI for Peace initiative under DeepFunding, and has successfully led campaigns around digital literacy, responsible tech use, and misinformation awareness earning a reputation for consistency, transparency, and values-driven leadership.
Ayomide Osemene hosts weekly X spaces for Cardano, titled CARDANO NEWCOMERS SUPPORT SPACE, she hosted the X space for several weeks, until her account got suspended, then she created another X account and continued hosting the space for the sake of onboarding people to the ecosystem
She is joined by Abdulrahman Abdulbasit Adigun, founder of Nextrend Labs, a forward-thinking technology and research company focused on innovation in Africa. Abdulrahman brings technical leadership, startup experience, and access to a network of developers and researchers essential for building the core AI components of PeaceGuard AI.
To ensure a high level of trust and accountability:
The team will share transparent monthly progress reports and milestone updates through Catalyst and social platforms.
A public pilot phase will involve journalists, educators, and peace advocates for real-time feedback and transparency.
All funds and expenditures will be tracked and reported clearly.
A final MVP demo and impact report will be publicly accessible.
Select on-chain integrations (such as user identity or truth proofs) are also planned for future phases.
Additional talents includes experts in at least two languages, UI/UX design, mobile/web development, and peace communication will be recruited as needed to ensure the project is executed efficiently and ethically.
Milestone Title
Planning & Team Onboarding
Milestone Outputs
Acceptance Criteria
Team roles formally documented; Kickoff meeting held; Tools and repositories created; Workflow agreed upon; All collaborators signed NDAs and agreements; Work begins based on roadmap.
The above will be the acceptance criteria
Evidence of Completion
we would provide signed agreements, Gantt chart/timeline, internal project documentation for all workdone.
Delivery Month
1
Cost
8500
Progress
10 %
Milestone Title
Dataset Collection & Curation ( at least two languages)
Milestone Outputs
Collection of misinformation-related datasets in two languages from reliable sources, including social media, news archives, and fact-checking platforms. Data will be cleaned and annotated for AI training.
Acceptance Criteria
The detection model must achieve at least 85% accuracy on a test set comprising real-world misinformation examples in each supported language; the language modules must correctly identify key semantics in the languages used; the system must flag misinformation in the three content modes (text, audio transcription, video metadata) with consistent results; test runs must complete under a fixed time and with acceptable false-positive rates; internal peer review of code and data must be completed.
Evidence of Completion
Model training and validation logs showing performance metrics and accuracy for each language; unit test reports for the NLP modules; example flagging outputs demonstrating detection on text, audio, and video; repository access with documented code, version tags, and test results; a recorded demo showing the engine in action, flagging sample inputs live; peer review notes or code review comments logged in GitHub.
Delivery Month
1
Cost
20000
Progress
20 %
Milestone Title
MVP Design & Platform Build
Milestone Outputs
Construction of a functional Minimum Viable Product accessible via both a lightweight Android mobile app and a responsive web interface. Users can submit text, voice, or video inquiries which are streamed to the detection engine and results (including confidence scores and supporting references) are displayed in the languages we are working with. The platform features a clean UI with multilingual labels, loading indicators, error handling, and anonymized analytics capture for usage tracking.
Acceptance Criteria
Demonstrable end-to-end functionality via both mobile and web front-ends, including multi-modal input methods and real-time verdict outputs; UI/UX must be responsive, accessible, and understandable to early users in local languages; system latency must remain under 5 seconds on average; three rounds of internal QA tests (functionality, performance, and UX) completed; an initial group of at least 10 testers from target user groups (educators, journalists, peacebuilders) can complete basic tasks without confusion in the languages.
Evidence of Completion
Screenshots of mobile and web interface flows in both languages; a demo video illustrating user input to output; GitHub release version for MVP with tagged commit; QA testing reports with test cases, results, and any bug resolution logs; usability feedback summaries from early testers; interface design documentation or style guide snapshot; performance logs showing system latency metrics.
Delivery Month
2
Cost
15000
Progress
50 %
Milestone Title
Community Testing & Feedback
Milestone Outputs
Conduct community beta testing with at least 50 real users drawn from educators, journalists, peacebuilding networks, and grassroots activists in targeted regions. Distribute structured feedback surveys and analytics to capture user satisfaction, detection accuracy in context, and mobile usability in low-bandwidth environments. Compile a comprehensive report mapping insights, feature suggestions, usability issues, and user retention stats.
Acceptance Criteria
At least 70% of testers indicate satisfactory experience (via Likert-scale responses); bug reports or improvement suggestions logged and triaged; language-specific usage patterns analyzed and documented; user behavior captured in metrics dashboards (session duration, click rates); incorporation of top 5 user-suggested UX or detection improvements before public launch.
Evidence of Completion
User feedback spreadsheet; screenshots of analytical dashboards with usage metrics; issue tracker logs for reported bugs; marked-up changelog showing resolved items; updated MVP version highlighting improvements based on community input; summary report detailing feedback methodology, sample quotes, and actionable recommendations.
Delivery Month
1
Cost
12500
Progress
70 %
Milestone Title
Public Launch & Monitoring
Milestone Outputs
Official public launch of PeaceGuard AI via published Android app (Google Play or APK distribution) and a live web portal. Launch includes awareness campaign materials (social media posts, local community alerts, press release). Support infrastructure includes a knowledge base with FAQs, user onboarding tutorial, and an issue-reporting system for real-time user feedback.
Acceptance Criteria
Minimum of 50 unique users access the platform in the first month post-launch; system records must show user engagement analytics such as user sessions, queries per language, and retention rates; help desk or issue reporting mechanism must capture all user issues; at least 80% of critical technical issues resolved within 2 business days; campaign reach metrics and community engagement recorded with screenshots.
Evidence of Completion
App store listing or web deployment link; analytics dashboard snapshots showing user counts and query data; screenshots of knowledge base and help desk; incident tracker logs with resolution times; copy or screenshots of campaign posts, press or media mentions; weekly progress summaries demonstrating platform stability and adoption trends
Delivery Month
2
Cost
15000
Progress
90 %
Milestone Title
Refinement, Reporting & Future Scaling
Milestone Outputs
Finalize product iterations including performance optimization, bug fixes, and UX polishing based on public feedback. Produce a comprehensive public impact report summarizing usage metrics, case studies, lessons learned, and the project's social value. Open-source the project under an MIT License (if approved), complete public documentation, and develop a robust roadmap for scaling, covering more languages, decentralized identity, on-chain verification, and sustainability strategies.
Acceptance Criteria
Detection accuracy emerges over 90% in real-world usage; final system performs robustly across all interfaces; open-source repository is published with full documentation, readme, and license; impact report shared publicly, with at least one community review event held; next-phase proposal and roadmap clearly defined and shared with stakeholders; Catalyst community acknowledgement of project success.
Evidence of Completion
Link or screenshot of published GitHub repository with MIT license; PDF/online version of the impact report; launch video or presentation recording; roadmapping document outlining Phase 2; community event recording or announcement; Catalyst feedback or social acknowledgment; performance logs evidencing robust system behavior.
Delivery Month
2
Cost
17500
Progress
100 %
Please provide a cost breakdown of the proposed work and resources
Milestone 1 – Initial Planning, Team Formation, and Research Setup (8,500 ADA)
Personnel (40%) – 3,400 ADA: Project lead, AI engineer, and research assistant stipends.
Research Tools & Resources (25%) – 2,125 ADA: Access to datasets, AI/NLP research journals, initial cloud computing credits.
Project Management Tools (15%) – 1,275 ADA: Subscriptions for Notion, Slack, GitHub, and Trello.
Meetings & Coordination (10%) – 850 ADA: Team kickoff workshop, coordination costs.
Contingency (10%) – 850 ADA: Buffer for unforeseen expenses
Milestone 2 – AI Model Development and Multilingual Dataset Training (20,000 ADA)
AI Engineering Personnel (45%) – 9,000 ADA: Senior AI developer, data scientist, and NLP specialist.
Cloud Computing & GPU Resources (30%) – 6,000 ADA: Model training and storage.
Dataset Acquisition & Annotation (15%) – 3,000 ADA: Yoruba, Igbo, and Hausa misinformation datasets, labeling costs.
Testing & Model Validation (5%) – 1,000 ADA: Internal accuracy checks and benchmarking.
Contingency (5%) – 1,000 ADA: Risk management buffer.
Milestone 3 – MVP Platform Design and Development (15,000 ADA)
Frontend & Backend Developers (40%) – 6,000 ADA: Web and mobile platform engineers.
UI/UX Design (20%) – 3,000 ADA: Multilingual interface creation, accessibility compliance.
API & Integration (15%) – 2,250 ADA: Linking AI detection engine with platform.
Quality Assurance Testing (15%) – 2,250 ADA: Functionality, usability, and security tests.
Contingency (10%) – 1,500 ADA: Technical risks.
Milestone 4 – Community Testing and Feedback Collection (12,500 ADA)
Community Outreach & Beta Recruitment (35%) – 4,375 ADA: Engaging 100+ testers in Nigeria and diaspora.
Platform Hosting & Scaling (25%) – 3,125 ADA: Increased server capacity for testing phase.
User Testing Coordination (20%) – 2,500 ADA: Surveys, interviews, analytics tools.
Bug Fixing & Improvements (10%) – 1,250 ADA: Quick iterations based on tester feedback.
Contingency (10%) – 1,250 ADA: Unexpected issues.
Milestone 5 – Public Launch and Awareness Campaign (15,000 ADA)
Marketing & Communications (40%) – 6,000 ADA: Social media ads, PR, and content creation.
Launch Event & Webinar (20%) – 3,000 ADA: Public demonstration and onboarding.
Customer Support Setup (15%) – 2,250 ADA: Help desk staffing and training.
Infrastructure Scaling (15%) – 2,250 ADA: Ensure smooth performance post-launch.
Contingency (10%) – 1,500 ADA: Buffer for operational needs.
Milestone 6 – Refinement, Reporting, and Future Scaling (17,500 ADA)
Platform Optimization (35%) – 6,125 ADA: Performance upgrades, bug fixes, and UI refinement.
Impact Assessment & Reporting (25%) – 4,375 ADA: Data analysis, community reports, and audit preparation.
Open Source Release (20%) – 3,500 ADA: Documentation, licensing, and code publishing.
Future Scaling Research (10%) – 1,750 ADA: Exploration for expansion to other African languages.
Contingency (10%) – 1,750 ADA: Reserve for last-stage adjustments.
How does the cost of the project represent value for the Cardano ecosystem?
This proposal demonstrates strong value for money by ensuring that each ADA allocated is tied directly to measurable outputs and tangible progress at each milestone. The budget distribution is milestone-based, preventing overspending and guaranteeing accountability.
Efficient Resource Allocation: Funds are proportionately assigned to key activities such as system development, AI model training, multilingual data collection, community engagement, and testing, ensuring no wasted expenditure.
Milestone-Based Payments: Payment is tied to deliverables across Milestones 1–6
Leveraging Open Source: By adopting an open-source framework, the project reduces future maintenance costs, enables community-driven enhancements, and avoids expensive proprietary licensing fees.
Long-Term Impact: The solution is designed for scalability across multiple conflict-affected regions, providing ongoing value beyond the project duration without requiring significant recurring costs.
Co-Funding Potential: The infrastructure and tools created will be reusable by other peace and misinformation detection initiatives, further multiplying the return on investment.
In short, the proposal ensures that Catalyst funding is used effectively to build a high-impact, sustainable, and reusable AI-based misinformation detection framework while maintaining transparency, accountability, and measurable outcomes.
Terms and Conditions:
Yes
Ayomide is a dynamic final-year university student, tech educator, and community builder with a passion for leveraging technology for education, development, and social impact. She has led and contributed to diverse projects.
The initiator of Helping ULSESAites, a student-led welfare project.
Her work spans education innovation, digital literacy, marketing, communications, and technology for development, with a focus on youth empowerment, online safety, and productivity tools.
Ayomide blends strategic thinking with hands-on execution, whether she’s designing community engagement calendars, managing digital campaigns, developing product concepts, or mentoring others.
Driven, creative, and solutions-oriented, she is committed to building scalable, impactful projects that bridge knowledge gaps, foster collaboration, and inspire others to explore the intersection of technology and positive change
She was the event planner/co-ordinator of the Nextrend Hub Launch in Lagos, Nigeria with partnership with WADA on the 16th of August, 2025. The event was all about mass adoption of Cardano and education.
Abdulrahman Abdulbasit Adigun is a Nigerian entrepreneur, blockchain advocate, and Al researcher. He is the founder and CEO of Next Trend Labs and Next Trend Group, which focus on using decentralized solutions to solve socio-economic and political issues in Nigeria and Africa.
Currently a final-year undergraduate at the University of Lagos, studying mathematics and education, Abdulrahman has a diverse professional background in product development, digital marketing, blockchain, and Al. He's an active member of the Deep Funding Marketing and Review Circles and has been deeply involved in the Cardano ecosystem since 2021, contributing to Project Catalyst, governance initiatives, and organizing DRep workshops in Nigeria.
Through Next Trend Labs, he leads projects that merge blockchain, Al, and community development, such as Nova, an Al tutor, and the CSP-E App, a security solution for seed phrases. His research focuses on integrating decentralized Al and loT systems for applications in smart homes, healthcare, and agriculture.
As a strong advocate for technology in Africa, Abdulrahman champions Nigerian involvement in global blockchain discussions, aiming to foster innovation, security, and sustainable growth.