EthicsNet Creed.Space

EthicsNet Creed.Space combines two complementary initiatives in the AI ethics and safety space, both led by engineer and philosopher Nell Watson. EthicsNet, founded around 2016, is a nonprofit community dedicated to co-creating datasets of prosocial behaviors for machine ethics algorithms. Its Massive Transformational Purpose is to 'Help Machines Make Decisions That Enable Flourishing.' Modeled after Fei-Fei Li's ImageNet project that transformed computer vision, EthicsNet aims to do the same for machine ethics by crowdsourcing annotated examples of prosocial behavior from people across different cultures, geographies, and value systems. The project recognizes that what constitutes prosocial behavior varies across these dimensions and seeks to democratize contributions to ethical AI development. EthicsNet provides a browser extension for Chrome and Edge that lets users annotate examples of behavior while browsing, and previously ran the Guardians' Challenge competition on HeroX with a $10,000 prize pool for ideas on generating prosocial datasets. Creed.Space is a more recent platform spawning from EthicsNet's work, focused on personalized constitutional AI alignment. The platform's core innovation is the Personalized Constitutionally-Aligned Agentic Superego framework, which functions as an oversight agent that monitors AI behavior in real time and checks outputs against user-defined ethical parameters. Users can select from a library of 'creed constitutions' representing specific value systems -- ranging from religious requirements to corporate safety standards, vegan principles, K-12 educational appropriateness, and fiduciary duties -- and set adherence levels on a 1-5 scale. The system integrates with GPT-4o, Claude, and Gemini via the Model Context Protocol. In published benchmarks (Information journal, Volume 16, Issue 8, 2025), the framework demonstrated a 96.4% reduction in jailbreak success rates on GPT-4o and a 76.9% reduction on Gemini 2.5 Flash, with a 99.4% refusal rate on harmful content. The research team includes Nell Watson alongside Shujun Zhang (University of Gloucestershire), Ahmed Amer, Evan Harris, and Preeti Ravindra. The EthicsNet core team includes approximately ten members spanning machine learning, information security, blockchain, cybersecurity, mathematics, ethics, and cultural technology. Notable advisors have included Prof. Wendell Wallach (ethics scholar), Prof. Amitai Etzioni (socio-economist), and Dr. Louis Rosenberg (CEO of Unanimous AI). The organization has received support from the Future of Life Institute, Survival and Flourishing Fund, NVIDIA Inception Program, SIDNfonds, AI Xprize, Y Combinator Startup School, the Foresight Institute, and others. It uses Players Philanthropy Fund as its fiscal sponsor.