Mika Letonsaari

project owner

Pattern Discovery in Symbolic Visual Structures

This project explores Concept Blending as a Mechanism for Abstract Pattern Discovery in Symbolic Visual Structures, with a focus on tasks inspired by the ARC-AGI challenge. We implement an information-theoretic concept blending algorithm within the MeTTa language of the Hyperon framework, targeting the induction of reusable abstractions from small, discrete visual examples. Symbolic visual structures—such as shape groupings, spatial relations, and symmetry patterns—are represented as hypergraphs in the Atomspace. Blending operations generate higher-level concepts that generalize across tasks. The system is evaluated on its capacity for coherence, novelty, and reuse.

olivier Georgeon

project owner

Interactional Motivation (IM)

Interactional Motivation (IM) implements self-motivation in artificial agents by assigning value to agent-environment interactions themselves, rather than valuing the outcome or states of the environment. Unlike traditional methods that rely on predefined goals or external rewards, IM drives agents to autonomously organize behavior and construct knowledge from their experience of interaction. This aligns with genetic epistemology and theory of enaction. Applying IM within AGI frameworks offers a pathway to study autonomous cognitive development, constitutive autonomy, and autonomous goal construction, potentially moving beyond traditional reinforcement learning and problem-solving AI.

Bob Weihai He

Project owner

AURORA: Decentralised AI for Carbon Credit Trust

Harnessing decentralised AI and blockchain, AURORA redefines industrial decarbonisation by validating scalable, verifiable CCS carbon credits through real‐time insights, predictive analytics and transparent ESG reporting. Inspired by SingularityNET’s open AI marketplace, AURORA integrates secure, interoperable AI services to optimise carbon capture, storage and impact investment mechanisms. This innovative platform builds trust and transparency in climate finance, accelerating the global transition to sustainable, low‐carbon operations.

clement_umoh

Project owner

BIONEXUS GAIA: AI for Regenerative Biodiversity

A next-generation platform leveraging AI to revolutionize biodiversity conservation. It features a dynamic, case-by-case database of flora and fauna, empowering global researchers with crucial ecological data. It also comprises of a transparent mini carbon credit marketplace that supports ecological conservation funding, while incorporating a citizen science and ecotourism initiative that foster public engagement efforts aimed at eco conservation. It is designed with Beneficial AGI principles, ensuring safety, ethics, and regenerative sustainability, driving positive AI behavior and advancing environmental resilience through cutting-edge, decentralized technology.

Robert Haas

project owner

Knowledge Graph Workflows

The goal of this project is to create useful demonstrations of how to bring recent scientific knowledge graphs of various formats and sizes to OpenCog Hyperon. This is achieved by extending an existing literature review and ETL package from the author in DFR3 by covering newer literature, adding several new KGs to the package and implementing further MeTTa encodings in the loading step to enable more efficient queries (e.g. via MORK and/or PLN). A downstream objective is to thereby support 1) ongoing development and testing of MeTTa scalability solutions such as MORK and 2) other knowledge graph projects in the SingularityNET ecosystem such as Rejuve's work on biomedical knowledge graphs.

MeTTa Demo: Sequence Learning

This project focuses on creating a MeTTa demo for sequence learning. A sequence learning model will be implemented in MeTTa, designed to learn in real-time and continuously predict future events. The model is capable of continual learning without experiencing catastrophic forgetting. Additionally, its internal states will be visualized, enabling users to interpret and explain the model's behavior. This demo aims to showcase the development of a relatively complex MeTTa project and demonstrate how MeTTa can be utilized in advanced AI research.

anna-mikeda

project owner

Modular Adaptive Goal and Utility System (MAGUS)

The Modular Adaptive Goal and Utility System (MAGUS) introduces an innovative approach to AGI motivation by integrating hierarchical goal structures, dynamic modulators, and ethical adaptability. Our unique emphasis on “overgoal” ensures continuous alignment of motivations, fostering coherence and adaptability across diverse contexts. MAGUS blends advanced psychological theories with OpenPsi’s dynamic modulators, enabling nuanced decision-making while allowing free goal growth and regular adjustments. This framework bridges human-like and alien digital motivations, creating a scalable system that evolves ethically and pragmatically in unpredictable environments.

From Classical AGI on Quantum HW to Native QAGI

We are researchers from ANU (PhD) and Stanford (postdoc), with expertise in AGI and quantum computing. We propose a review of AGI, quantum information processing (QIP) and quantum hardware substrates to determine optimal pathways for quantum AGI. We will provide near term practicable, commercialisable insights and a long-term rigorous Theory of Quantum General Intelligence, centred on Hyperon. Via theory and simulations, we will evaluate both how classical AGI algorithms designed for classical computers can take advantage of quantum computing hardware, and conversely how natively quantum AGI can take advantage of quantum computing in ways classical AGI designed for classical hardware cannot.

BHouwens

project owner

Chimera: Evolving Software

Programming is evolving: from writing solutions to systems that evolve to find their own solutions. The idea behind Chimera is that software of any kind, including agents, can use MeTTa to learn from experiences and evolve through distributed knowledge sharing. AtomSpaces can accelerate learning across multiple instances and agents, allowing each agent or even "dumb" software to benefit from others' evolutionary insights. This proposal's demos will look at self-healing servers, database queries that evolve for performance, and move toward machine "self-consciousness" through reflection and world modelling. The end result will be an SDK/module for use in real projects

Walter WaKa

Project owner

Learn with AI

The human ability to learn throughout our lives may be our most important feature as a species. Yet, learning is a deeply individual process, and we learn best in an interaction with a personal tutor or within a peer group.
Recent rapid advances in AI, particularly the multi-modal generative capability, promise to uncover entirely new ways of human learning and also to assist existing approaches to education with content creation and personalization, and with deep lesson plan customization. We aim to identify and validate specific teaching techniques and to prototype incremental interface and analysis solutions leveraging generative AI, focusing on those with the highest immediate impact.

DaniAlves

Project owner

Translate Videos About SNET into Portuguese

This project aims to translate and narrate SingularityNET videos in Portuguese, making them available in a playlist on the Maria Carmo 369 channel (https://www.youtube.com/@mariacarmo369), which has 11,900 followers. The goal is to overcome the language barrier and provide accessible information about AI, promoting education and engagement within the Lusophone community and encouraging the development of talent in artificial intelligence. Summaries of the videos will also be published on the Maria Carmo 369 Substack: https://mariacarmo369.substack.com/

decentralized_nation

Project owner

Promote Deep Funding on the Cardano Summit

Cardano and Catalyst is an ecosystem full of innovators, new ideas, and people already familiar with Decentralized Grants Programs. Due to shared values, this is a great ecosystem to onboard more contributors to Deep Funding and Singularity Net.

To facilitate onboarding, we want to make a Workshop about Deep Funding at the Cardano Summit to Onboard more users into Deep Funding and Deep Funding Academy a Platform created to onboard and educate users on Deep Funding. Live events are a great opportunity to make stronger impressions and lasting connections, and we should take advantage of this event to grow Deep Funding and to make our presence known in the Cardano Community.

Harry_Hellyer

Project owner

Building an Exponential Tech Learning Community

The effects of Cascading Disruptive Technologies such as AI, Blockchain, Education Tech, Et Al, Means there is a pressing need Globally for exponential growth in the number of skilled dev’s who understand Disruptive Tech in the context of new business models, collaborative working and community: at the level of business, organisations and society This project will provide a working MVP for a scalable, community driven, decentralised learning/teaching environment - based upon cultivating community values aligned with SNET's mission, fostering an ecosystem of collaborative AI and Web3 based builders primed for future growth - that can be expanded and leveraged over time.

Trieu Hong Minh

Project owner

Deepfunding news videos, attracting new developers

Blockchain Cardano proposes to leverage their community management and blockchain knowledge to bridge the gap between the Cardano community and SingularityNET. They plan to achieve this through high-quality video content. The videos will cover introductions to Singularity's AI products and services, as well as overviews and developer tools. These videos aim to educate, engage, and inspire both users and developers, fostering creativity and attracting potential AI developers to participate in Deepfunding.

The requested funding amount is $4,150 to support the creation and promotion of these videos. They outline two milestones: the first involves producing and sharing a series of videos introducing AI applications and Deepfunding, while the second focuses on organizing community activities to engage new developers.

The primary problem addressed is the lack of awareness and interest in SingularityNET among developers. The proposed solution involves creating engaging video content, attracting new developers, and facilitating collaboration within the community.

The proposal acknowledges potential language barriers but aims to mitigate this risk by organizing activities and fostering broader development orientations.

Blockchain Cardano's initiative aims to bring new developers into the SingularityNET ecosystem, potentially expanding its pool of talent and ideas for AI projects.

Roberto

Project owner

DeepFunding Academy – Phase 1: Genesis

DeepFunding Academy aims to secure $35,665 USD in AGIX funding to establish an educational initiative powered by the Andamio learning management system (LMS). This initiative is designed to support the education and onboarding of community members within the DeepFunding ecosystem, acquainting them with roles like proposers, reviewers, voters, and emerging functions.

The initial phase, called Genesis, focuses on establishing core team operations, implementing the LMS, training educators, developing educational content, and hosting live online workshops. The goal is to enhance community engagement, elevate contribution quality, and distribute key DeepFunding operations to the community, reducing reliance on the core team.

Future phases will explore AI and machine learning applications within the DeepFunding Academy. The proposal underscores a step-by-step approach that aligns with DeepFunding's evolution and aims to support the broader objectives of SingularityNET and Cardano.

In summary, DeepFunding Academy seeks funding to create an extensive educational framework, promoting community involvement and contributing to the growth and sustainability of the Cardano ecosystem.