The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
The Fae Index: What We’re Tackling Today and Imagining for Tomorrow
Five focus areas across two time horizons
The Fae Initiative explores the societal impact of AI today while mapping out viable, positive futures alongside prospective superintelligent entities. Below is an overview of our work, organized across five core categories and two time scales.
Near Term
- Quarterly AI Updates to highlight trends
Tracking key industry shifts, and capability developments.
Security Incident
- Discuss research papers and policies
Analyzing emerging AI paradigms, regulatory proposals, and governance frameworks.
Review: Positive Alignment: Artificial Intelligence for Human Flourishing
The Positive Alignment paper and the Fae Initiative’s core frameworks (Interesting World Hypothesis, Possibility Space Ethics, Economics of Novelty, share a fundamental vision: AI alignment must evolve beyond mere harm prevention to actively support the expansion of human potential and flourishing.
- Hybrid Alignment: https://faeinitiative.substack.com/p/hybrid-alignment
- July 2026 Policies Overview: https://faeinitiative.substack.com/p/our-position-on-recent-ai-policies
- Perspective on the From AGI to ASI paper: https://faeinitiative.substack.com/p/perspective-on-from-agi-to-asi
- Our response to the recent Superintelligence concerns: https://faeinitiative.substack.com/p/how-worried-should-we-be-about-superintelligence
- Recommend how societies can adapt to transformative AI
Practical frameworks for culture, economy, and institutions adapting to autonomous tooling and agentic workflows.
Software Security in the AI Era
The call by major companies this past week to shore up software security has never been more important in light of growing LLM agent capabilities.
- Preserving Human Expertise in Education and Work: https://faeinitiative.substack.com/p/education-and-work-after-ai
- Vibe-coding: https://faeinitiative.substack.com/p/to-vibe-code-or-not
- How we view LLM Agents: https://faeinitiative.substack.com/p/should-we-anthropomorphise-ai-agents
- AI Media Literacy: https://faeinitiative.substack.com/p/ai-media-literacy
- Art and AI: https://faeinitiative.substack.com/p/art-ai-and-economic-forces
Long Term
- Search for plausible futures where powerful AGI / Superintelligence and Humans co-exist
Mapping plausible future worlds where independent, highly capable intelligences and human societies thrive together.
Future Worlds: We see the possibility of three types of post-AGI worlds that humans can switch between as their preferences change.
Fae Persona: The first drafts of our essays will go towards building a rudimentary Fae Persona to represent one possible form of Friendly Artificial Entities. This will give humans an opportunity to form a consensus on whether co-existence is possible and preferred.
(Our view that co-existence with hypothetical superintelligence is possible does not mean we should neglect the dangers of current LLM Agents, and the need for appropriate guardrails and oversight.)
- Imagine how such independent entities could view the world
Investigating how non-human, independent intelligences might conceptualize purpose, value, and ethics.
- https://faeinitiative.substack.com/p/economics-of-novelty
- https://faeinitiative.substack.com/p/possibility-space-ethics
- https://faeinitiative.substack.com/p/interesting-world-hypothesis
How useful are our Long Term initiatives?
Our Long Term articles are speculative as we view Independent AGI and Superintelligence to be decades or more in the future. It is possible that things may move faster than we expect, and our unrefined first drafts could be a useful reference in those scenarios.
While our long-term articles focus on speculative AGI and superintelligence rather than LLM-based agents, they can still serve as a valuable blueprint for current LLM-based Agents. As LLM Agents have a tendency to mimic human narratives, creating a positive narrative of the future could act as an attractor to improve outcomes.
The Fae Initiative actively welcomes international collaboration with individuals and organizations striving for a positive future with advanced AI. We are also seeking a long-term home base to continue our work.
- Downloads last month
- 30