{"slug": "some-research-and-project-ideas-for-developing-and-refining-the-field-of-ai", "title": "Some research and project ideas for developing and refining the field of AI safety", "summary": "A researcher proposes several high-impact AI safety projects, including live databases of research questions and interventions to track progress and neglectedness, and exploring under-contacted disciplines and attractor basins as risk vectors.", "body_md": "This small post is part of a series of posts I'm creating which outline my ideas for AI safety projects/research which I think could be high-impact. I have way more ideas than I could ever hope to do justice to, so if you're looking for inspiration, take a look!\n\n**There should be a live, regularly-updated and highly visual database of AIS research questions and the progress on each of them***.* The nearest existing thing is the [EA Midwest list-of-lists](https://www.notion.so/eamidwest2026/List-of-lists-of-projects-ideas-in-AI-Safety-291f040b30f180a794c3f1f6b5264534); it could be far more. Could plausibly be made and maintained mostly with AI scraping and auto-categorisation. A database like this would allow people to easily see the state of research on a given subtopic, and where there is progress to be made. I suspect that the majority of updating could be automated.\n\n**Similarly, there should be a live, regularly-updated and highly visual database of proposed interventions in AI safety, which tracks how many people are working on each intervention, and roughly how much time they're on each intervention***. *This would allow us to more quantitatively assess neglectedness.\n\n**Which disciplines have had the least contact with AI safety, and what could they contribute?** See this Claude deep research artefact: [https://claude.ai/public/artifacts/55bde7d3-2216-43ea-83f0-9857e1e48750](https://claude.ai/public/artifacts/55bde7d3-2216-43ea-83f0-9857e1e48750)\n\n**Are attractor basins from AI use a risk vector that curtails genuine innovation in AIS itself?** One candidate antidote: deliberately spending time in layers of reality that are far from AI (eg nature) in order to tap into sources of inspiration which lack these attractor basins.\n\n**Given the high degree of disagreement among experts regarding which AI safety interventions are most promising, would it be helpful for intervention comparisons to factor in:**", "url": "https://wpnews.pro/news/some-research-and-project-ideas-for-developing-and-refining-the-field-of-ai", "canonical_source": "https://forum.effectivealtruism.org/posts/Tgj8N6Rh4HHBAAyav/some-research-and-project-ideas-for-developing-and-refining", "published_at": "2026-06-24 18:23:22+00:00", "updated_at": "2026-06-24 18:44:06.553406+00:00", "lang": "en", "topics": ["ai-safety", "ai-research", "ai-policy"], "entities": ["EA Midwest"], "alternates": {"html": "https://wpnews.pro/news/some-research-and-project-ideas-for-developing-and-refining-the-field-of-ai", "markdown": "https://wpnews.pro/news/some-research-and-project-ideas-for-developing-and-refining-the-field-of-ai.md", "text": "https://wpnews.pro/news/some-research-and-project-ideas-for-developing-and-refining-the-field-of-ai.txt", "jsonld": "https://wpnews.pro/news/some-research-and-project-ideas-for-developing-and-refining-the-field-of-ai.jsonld"}}