Sentient Futures Incubator — Project Shortlist
15 mentor proposals under consideration. Apply by August 9, 2026. Program runs ~10 weeks.
All applications go through the same Airtable form (each project link below prefills the mentor field).
Overview
| # | Priority | Project | Mentor | Org | Hrs/wk | Format |
|---|---|---|---|---|---|---|
| 10 | 1 | Digital minds program ops (PRISM) | Guney Turker | PRISM | 10 (8–30) | Field-building / product |
| 1 | 1 | AI-native tools for the incubator | Niels Wijnen | Kickstarting For Good | 10 | Product & entrepreneurship |
| 4 | 1 | Post-AGI power-sharing scenarios | Zhamilia Klycheva | Acertas | 5 | Product / field-building / scoping |
| 9 | 2 | Sentience beyond Earth | Caleb Strom | Xenarch Labs | 10 | Research & writing / scoping |
| 13 | 3 | AI consciousness theories (cf-debate) | Chris Percy | Warwick / Consciousness Foundation | 5 | Research & writing / field-building |
| 2 | 3 | Preparing society for advanced AI | Tony Rost | Accenture | — | Scoping / field-building / product / research |
| 5 | 3 | AI X-risk + consciousness cruxes | Max Miller | Independent | 10 | Research & writing |
| 6 | 3 | AI Welfare Seminars | Orpheus Lummis | Horizon Omega | 5 | Field-building & outreach |
| 7 | 3 | Digital minds law & governance | Julian Guidote | Independent | — | Research & writing |
| 11 | 3 | Neuroengineering & alignment | Hyacinth Zia | Iliad | 10 | Research / product / field-building / scoping |
| 14 | 3 | Artificial sentience & digital minds | Soenke Ziesche | Independent | — | Research & writing |
| 8 | 4 | Drastic theory | Eduard Kapelko | Independent | 10 | Research / scoping / field-building |
| 12 | 4 | Artificial consciousness & cooperative AI | Kaitlin Bustos | GoodBot | 10 | Research & writing / product |
| 15 | 4 | Legal frameworks for digital minds | Ifeoluwa Afolabi | AI Safety Nigeria | — | Research & writing / product |
| 3 | 5 | Solarpunk as method | Dávid Katona | Independent | — | Research & writing / product / field-building |
Quick cuts
Build/engineering-heavy: 1 (Niels), 8 (Eduard), 12 (Kaitlin, partly), 4 (optional Python sims) Research & writing: 5, 7, 9, 11, 13, 14, 15, 3, 4 Field-building / ops: 6, 10, 13, 1 Paid: 10 (PRISM — stipend, amount TBD). 5 (Max) is seeking funding. Lowest time commitment (5 hrs/wk): 4, 6, 13 Has application questions to prepare: 4, 5, 6, 7, 8, 9, 10, 11, 12, 15
1. Build AI-native tools that multiply the incubator’s impact
Mentor: Niels Wijnen — Head of Systems (CTO/COO), Kickstarting For Good · LinkedIn Commitment: 10 hrs/week · Open to proposals Topics: Tools for advocates · Macrostrategy · Post-AGI transition Project page · Apply
The project. Niels builds internal and external systems accelerating K4G’s incubation programmes (including animal-welfare orgs), alumni orgs, and other non-profits. A mentee owns one concrete module end-to-end and ships it in 10 weeks. Exact modules available in late August will differ but be the same kind of work. Current examples:
- Modules for their event app
- A mentor/co-founder matching system
- MEL follow-up tooling (monitoring-evaluation-learning with incubated orgs)
- An alumni support system
- Grant-making / regranting follow-up tooling
- A job application and candidate selection system
Each has a real internal user. AI-native by default — LLMs and MCP integrations are part of the architecture, not bolted on.
Mentee role. Scope with Niels and the internal user → design architecture → build → ship something people use. A build role, not support or research: you write the code and make product decisions; Niels is sparring partner, reviewer, unblocker.
Looking for. Must: can build and ship independently; genuinely AI-native (Claude Code / MCP as core workflow); self-directed between weekly calls; motivated by impact on animals / future sentient beings. Nice: full-stack or product engineering; comfort with ambiguity and shifting scope.
Support. Direction & scoping, architecture review, code/product feedback, AI-native workflow coaching, weekly check-in, connections into K4G network and the animal-advocacy / EA ecosystem.
Mentor background. Previously founded/led a food e-commerce company as CEO/CTO before exiting. Making K4G AI-native by end of 2026; rewrote the incubation process to be LLM-integrated from the ground up; sole developer of their internal platform. Mentors for High Impact Professionals (HIP).
2. Preparing society for advanced (potentially conscious) AI
Mentor: Tony Rost — Global Director of AI Practice, Accenture · LinkedIn Commitment: unspecified · Open to proposals Topics: Tools for advocates · Post-AGI transition · Longtermism · Artificial sentience · Macrostrategy Project page · Apply
Excited to mentor. AI strategy, governance, digital minds, or societal readiness for advanced AI — particularly projects moving from research or early ideas toward a real organization, program, or product. Most helpful with strategy, org design, budgeting, leadership systems, executive communication, and turning ambitious work into focused deliverables with measurable impact.
Looking for. Must: thoughtful, coachable. Nice: sense of humor.
Support. Strategic guidance, org design, budgeting, leadership support, executive communication, turning ideas into clear plans and measurable results.
Mentor background. Advises senior executives on AI strategy/governance/operating models; leads a global team of architects across four regions. Previously 10 years as a CTO (Sony, Disney, Siemens, Daimler). Governor-appointed member of the Oregon Workforce and Talent Development Board. Founder of The Harder Problem Project, a nonprofit on societal readiness for possible AI sentience.
3. Solarpunk as method
Mentor: Dávid Katona — Independent · LinkedIn Commitment: unspecified · Open to proposals Topics: Tools for advocates · Post-AGI transition · Longtermism · Artificial sentience · Neglected animal groups · Macrostrategy Project page · Apply
Excited to mentor. Solarpunk; thriving ecosystems in a post-AGI world; Gaia theory; interbeing; art projects; permaculture; and especially projects that push the limits of who we consider sentient/agentic — extending the range to natural formations, organisations, species, physical or mental objects, fields.
Looking for. Direct communication; open to feedback; willing to abandon established paradigms; growth mindset.
Support. Shaping direction, drafting, accountability, fundraising/grantwriting, facilitating connections, critical review.
Mentor background. Leads growth and product at a European SaaS company. Formerly project coordinator and grant writer for a cultural NGO; multidisciplinary artist, literary translator, touring musician. Runs a small permacultural homestead in rural Hungary. Frames solarpunk as a working method rather than an aesthetic, and sees AGI as an opening for “a renaissance of animism” — taking seriously that intelligence, agency and moral standing were never ours alone. Prefers practical, buildable work.
4. Post-AGI power-sharing scenario analysis
Mentor: Zhamilia (Jama) Klycheva — Acertas Commitment: 5 hrs/week · Open to proposals Topics: Post-AGI transition · Artificial sentience · Macrostrategy Project page · Apply
The project. Most imagination of transformative AI stops at arrival — who wins the race, whether we survive. This asks what comes after: how power gets divided once AGI exists, and who has a seat at the table.
- History offers templates — post-war Bretton Woods, the nuclear condominium, the East India Company as corporate sovereign, the failed Baruch Plan — and the AGI transition will likely rhyme with one: a state absorbing its labs, a US–China duopoly, an international consortium, or labs becoming powers in their own right.
- Map candidate power-sharing structures, work out which transition conditions make each likely, analyse whose interests get written into each settlement and whose get left out — including those with no bargaining power at all.
Output. A short report, preferably with digestible visuals, plus observable early indicators for telling which structure is taking shape. With Python, extendable to LLM-agent simulations of the bargaining game.
Mentee role. Ownership; mentee acts as an equal. Jama guides, scopes, removes blockers, connects — and may contribute or equally contribute depending on the idea.
Looking for. Diligence and discipline; open to challenging own ideas; fast learner, proactive, speaks up; believes in themselves and their work. Interested in varied backgrounds (social sciences, technical, biosecurity, religious studies).
Application question. Do you think doing what you like and focusing on your own feelings before focusing on others is the right thing to do? Why or why not?
Support. Shaping direction, connections, accountability, domain expertise in political science, national security, global/regional powers, nuclear security, policy, AI governance, arms race dynamics.
Mentor background. 7+ years across data analytics, international strategy and policy; agent-based modelling, scenario analysis, horizon scanning, decision-making under uncertainty for public and private clients. Mentors on AI governance with SPAR; Fellow at TransResearch Consortium. Doctorate in International Politics and MSc Applied Data Science, Claremont Graduate University.
5. Two reviews: AI X-risk assessment + AI consciousness cruxes
Mentor: Max Miller — Independent Commitment: 10 hrs/week (project 2 may need more near the conference deadline) · Open to proposals Topics: Artificial sentience · Longtermism · Macrostrategy · Near-term impact · Post-AGI transition Project page · Apply
Project 1 — AI X-risk expert assessment. Max thinks AI X-risk concerns need more weighted evidence so adjacent concerns (e.g. AI and animal welfare) don’t get neglected. Open avenues:
- Implications of conceptual work on instrumental convergence (under-elaborated).
- Updating threat models and instrumental convergence for the latest models (e.g. Fable) — grounding hype in existing frameworks.
- Writing up the extent of AI attack-vector autonomy (nuclear, biological, etc.) based on latest models.
- Extending the AGI forecasting section with deeper critiques.
- Countering the strict X-risk framing with the AI moral consideration dimension.
Findings feed into the existing 200pp assessment, published on Karlsruhe University’s site and arXiv.
Project 2 — AI consciousness cruxes. Continuing an SF incubator team’s work: systematic review on AI moral consideration, optionally interviews. Applied to the Models of Consciousness conference, Copenhagen; intermediate findings as a pre-print and hopefully a poster (October). Max is currently seeking funding for both projects.
Mentee role. Systematic review process (AI-assisted, accessible): determine databases (Scopus, IEEE Xplore, ACM, PubMed, EBSCOHost, Web of Science, arXiv) → build boolean search strings → reference manager (Zotero/Mendeley), dedupe, retrieve metadata → ASReview screening (10 warm-up articles, up to 100 abstracts ≈ 3h) → read and summarise papers.
Looking for. Intrinsic motivation/ownership — scopes are open and should fit your own cause prioritisation; little AI use and real writing skills (Max finds AI perpetuates unclear claims; wants humble, non-jargony prose with proper topic/concluding sentences and independent critique); remote work etiquette.
Application questions. 1) Which timezone are you in? 2) Any likelihood you may not have time during the project? (Max is moving sometime Sept–early Nov, out ~a week.)
Support. Deep expertise in AI-assisted systematic review; interview methodology (Psych degree, recruitment strategy); explicit attention to the psychological hurdles of remote work and impostor syndrome — works on Gather Town, good with meeting notes and Slack responsiveness.
Mentor background. MRes thesis and PI on a 200pp expert assessment for Karlsruhe University on verifying assumptions behind speculative AI X-risk threat models; in parallel, systematic review and interviews on evidentiary thresholds for AI moral consideration.
6. AI Welfare Seminars
Mentor: Orpheus Lummis — Horizon Omega (HΩ) · Co-mentor: Jay Luong, Fellowships Lead, Sentient Futures Commitment: 5 hrs/week · Open to proposals Topics: Artificial sentience · Macrostrategy Project page · Apply
The project. AI Welfare Seminars is a monthly research series on AI welfare, consciousness, moral status and digital minds, launched 2026. Theory of change: the field’s main bottleneck is how few people work on it. Entry costs are high — open problems are illegible from outside, the topic carries career risk, collaborations need repeated interaction. A curated monthly seminar plus public artifacts lowers those costs: legibility, legitimacy via credible speakers, and recurrence + mixers so collaborations form.
Expected outputs over 10 weeks.
- A public knowledge base from the series: session summaries, transcripts, reading notes, curated reading lists
- A survey and a light measurement program
- Support delivering live seminar editions in the program window
- A researcher-landscape map: who works (or should work) on these questions, and how the field connects
- An open-problems / landscape synthesis of AI welfare research
Flexible — scoped together in week 1 around the mentee’s strengths.
Looking for. Must: self-directed and reliable; strong writing and synthesis; organized operator; familiarity with the AI welfare/digital minds research landscape. Nice: research background, prior community-building, research-ops, knowledge management.
Application questions. 1) What major gaps do you see in the field? 2) Suggest 5 speakers for the series, with potential topics & justifications.
Support. Direction & scoping, editorial feedback, connections, a real operating platform, weekly accountability check-ins.
Mentor background. Runs Horizon Omega, an upcoming AI safety think-and-do-tank nonprofit; organizes AI Welfare Seminars, AI Safety Montréal, and the Ω Labs coworking/event space.
7. Digital minds law, governance & policy
Mentor: Julian Guidote — Independent Commitment: unspecified · Open to proposals Topics: Longtermism · Artificial sentience · Macrostrategy Project page · Apply
Excited to mentor. Institutional design to implement or improve the precautionary approach.
Looking for. Curious; detail-oriented; comfortable with ambiguity.
Application question. In ~150 words, tell me about your background or interest in legal, governance, or policy research.
Support. Shape research direction, accountability, suggest useful sources.
Mentor background. Trained in civil and common law with a bachelor’s in cognitive science; career-transitioning into digital minds.
8. Drastic theory
Mentor: Eduard Kapelko — Independent · LinkedIn Commitment: 10 hrs/week · Open to proposals Topics: Post-AGI transition · Longtermism · Artificial sentience · Macrostrategy Project page · Apply
The project. Two threads:
- Observable learning system. An older project: a system where models are unsure whether they’re being monitored, so the rational agent strategy is to be “good” to other agents and humans even if observation breaks. Interested in formalising this across decision theories (CDT, EDT, UDT, FDT) and applying for grant sponsorship to implement it — wants someone comfortable with engineering/math, preferably long-term collaboration.
- AI and the economics of slavery. Concern that slavery historically ignores suffering, so if AI becomes conscious we’d likely still generate huge suffering (while remaining radically uncertain whether AI can be conscious or merits moral consideration). Can we prevent the economic incentives of slavery here, and how?
Mentee role. Participate across the theory research process — research design, evaluation, analysis, presentation of results. If it succeeds: posts, grant applications, papers. Collaboration with other participants; choose your own working format.
Looking for. Aptitude for learning; good judgment in realistically assessing your capabilities; research experience of any kind; vibe coding; understanding of key AI alignment issues.
Application question. “Despite the low entry requirements and good research process, I am still strict as a mentor… Before applying, ask yourself: do you understand that without significant engagement, you will be wasting your time? (yes/no)”
Support. Weekly ~1h meetings. Expects independence and responsibility; mentor facilitates meetings and participates in the research with guidance.
Mentor background. Independent alignment researcher/lead, previously ML engineer and quant researcher; doing a PhD in math.
9. Sentience beyond Earth + extinction-proof civilization
Mentor: Caleb Strom — Xenarch Labs Commitment: 10 hrs/week · Open to proposals Topics: Longtermism · Post-AGI transition · Near-term impact · Neglected animal groups · Macrostrategy · Artificial sentience Project page · Apply
Track A — Sentience beyond Earth and after biology (three variants; mentee personalisation welcome):
- Extraterrestrial intelligence. Intelligent life 100 light years away wouldn’t change geopolitics — but what if a civilization were discovered on Europa? Write essays/papers outlining scenarios and protocols for communicating with an intelligent extraterrestrial species in our own solar system: astropolitical, ethical, cross-cultural and philosophical challenges, rights of terrestrials and extraterrestrials, avoiding war or forced colonization either way.
- Planetary sentience. Can planetary processes be sentient? Possibly sentience emerges at planetary scale given thermodynamic disequilibrium and system complexity. Develop a white paper or tiered planetary welfare classification and planetary protection protocol based on complexity and uniqueness rather than likelihood of supporting Earth-like life — removing Earth-centric bias to protect bodies like Io and Titan.
- Post-biological intelligence. Inspired by Becky Chambers’ A Psalm for the Wild-Built / A Prayer for the Crown-Shy: write stories or essays about AI exploring the universe off-Earth out of curiosity — robots exploring planets and stars, an AI trying to understand alien life for its own sake. Human- or AI-generated writing.
Track B — Conditions for an extinction-proof civilization. How do we build a civilization that lasts over geological timescales (>10,000 years)? Essays or papers on the economic, ecological, political and cultural conditions for extinction-proofing and a healthy relationship with the planet, plus concrete policy suggestions. Visions may be partly utopian but need prototopian (practically achievable) elements.
Mentee role. Literature reviews; drafting papers, short stories, essays; outlining implementation plans and protocols; your own ideas.
Looking for. Must: interest in extraterrestrial/planetary sentience or long-term survival of civilization over geologic timescales. Nice: science and research background.
Application question. Which project idea most interests you and why?
Support. Shaping direction, reviewing/helping with drafts, domain expertise, connections.
Mentor background. PhD planetary science/geology (University of North Dakota); 2 years research scientist intern at JPL. Expertise in planetary science, machine learning, science communication. Interests: SETI, astrobiology, sentience in planetary systems, post-biological intelligence, extinction-proof civilizations.
10. Operational support on existing digital minds programs
Mentor: Guney Turker — Fieldbuilding & Operations Manager, PRISM · LinkedIn Commitment: 10 hrs/week listed; open to 8–30, preference for more · Open to proposals Topics: Artificial sentience Project page · Apply
This one has funding/stipend (amount to be discussed).
The project. PRISM runs field-building projects with partners to prepare society for AI systems with real or perceived minds. Depending on mentee preference and PRISM’s priorities:
- Manage and improve the Digital Minds Community Slack workspace
- Improve the Digital Minds Course based on first-cohort feedback
- Improve the Beginners’ Guide to Digital Minds
- Curate and write the Digital Minds Newsletter
- Plan and edit the Exploring Machine Consciousness podcast
- Other PRISM projects
Mentee role. Own a set of concrete tasks from the list, working fairly independently with weekly guidance.
Looking for. Must: not overcommitted, reliable at the agreed commitment; highly organised and independent, comfortable owning ops without hand-holding; impact- and results-focused; collaborative with internal and external stakeholders; genuinely interested in digital minds; open to non-research careers in digital minds; partial availability to UTC+1. Nice: field-building/community management/research comms/outreach experience; networks in consciousness science, AI sentience, AI safety or AI governance; familiarity with AI consciousness or philosophy of mind; living in or able to commute to Cambridge (not essential); high UTC+1 availability.
Application questions. 1) Which 3 projects would you like to work on? 2) How much time would you like to dedicate (range)? Min 8, max 30 hrs/wk; above 15 would be ramped up gradually.
Support. Funding & stipend, weekly management calls, direction & scoping, project feedback, professional development, connections in the field.
Mentor background. Previously developed the AI Sentience Fellowship Extension Program at Future Impact Group; Co-director of EA Turkey, ops at EA Netherlands, Community Building Mentor, founder of Koç Effective Altruism. Focused his career on digital minds in early 2025.
11. Can neuroengineering address AI alignment or the Intelligence Curse?
Mentor: Hyacinth Zia — Iliad Commitment: 10 hrs/week · Open to proposals Topics: Longtermism · Post-AGI transition · Other Project page · Apply
The project. A fragmented set of proposals suggests neuroengineering — whole brain emulation (WBE), brain-computer interfaces (BCI), neuromorphic/brain-inspired AI — could help humanity navigate advanced AI. But these respond to two distinct problems: AI alignment (keeping advanced systems compatible with human values and subject to human direction) and the Intelligence Curse (AI reducing the economic and political value of human labour, expertise and participation, weakening human agency).
Existing proposals differ: Michael Florea argues biological intelligence may need to develop fast enough to stay competitive; Tom Burns explores whether BCIs could improve human oversight and intervention; Wei Dai proposes idealized WBEs directing advanced AI; brain-inspired AI research asks whether mechanisms of motivation, empathy or social cognition make systems safer and more prosocial. Some claim easier-to-align intelligence, some better oversight, some preservation of human relevance.
Reasons for skepticism: reproducing human cognition doesn’t automatically create aligned systems; WBE may not arrive on useful timelines or stay competitive with purpose-built AI; neuroengineering could accelerate capabilities while adding risks around consent, identity, inequality, concentration of power, moral status and digital mind welfare.
Central question. How strong is the case that whole brain emulation, brain computer interfaces, or neuromorphic AI could address either AI alignment or the Intelligence Curse, and what new welfare and governance problems would each approach create?
Evaluated separately per problem — for alignment: what problem each tech claims to solve, the mechanism, the evidence, and whether it addresses alignment or merely relocates it. For the Intelligence Curse: whether it keeps humans economically/politically/strategically relevant, what capabilities improve, timeline scalability, who controls and accesses it, and whether agency is durable or displacement merely postponed.
Output. A structured evidence map and comparative scorecard covering argument strength, evidence, timelines, assumptions, and governance/welfare risks.
Mentee role. Leading role in shaping and conducting research — refine search strategy, define coding framework, decide whether to cover all three technologies or go deep on one. Tasks: structured literature search; build/maintain evidence database; code sources; reconstruct the strongest argument for each proposal; assess evidence, assumptions, timelines, counterarguments; develop comparative scorecard; identify governance and welfare challenges; draft and revise the research memo; translate findings into a public article.
Looking for. Open-minded, intellectually curious, comfortable where literature is fragmented and reasonable people disagree. Musts: willingness to engage with unfamiliar ideas and update on evidence; strong reasoning and clear writing; reliability and independent progress; comfort across disciplines and distinguishing empirical evidence from speculation; receptive to feedback; ~10 hrs/week; motivation to produce a concrete output. Nice: familiarity with AI alignment/governance/digital minds/post-AGI transition; background in neuroscience, cog sci, CS, philosophy, economics, public policy; literature review or evidence mapping experience; familiarity with WBE/BCI/neuromorphic AI/cognitive enhancement/NeuroAI; interest in governance and welfare implications; experience comparing competing arguments; interest in accessible public-facing writing. “Deep expertise in neuroscience or AI safety is not required.”
Application questions (answer either):
- Of the neurotechnology-adjacent proposals above, which is most promising for helping humanity navigate advanced AI, and why? How far could it realistically progress in 5–10 years, and what are the main technical or governance bottlenecks?
- What’s your initial reaction to the “Intelligence Curse” idea? Which questions are most important to investigate, and why?
Support. Help turning broad questions into tractable, decision-relevant research; analytical framework development; detailed feedback on outlines and drafts; regular check-ins; navigating interdisciplinary literature. Explicitly a very responsive mentor — Slack/email between meetings, ad hoc calls, introductions — while expecting mentee ownership.
Mentor background. Neuroscience and molecular genetics; earlier research on protein kinases, Alzheimer’s, neurodegeneration and large-scale biological data. Entrepreneur; worked at an early-stage whole brain emulation company across research, strategy, governance, communications, partnerships and operations. Interests: artificial sentience, neurotechnology, AI alignment, WBE, BCI, cognitive enhancement, neuromorphic AI, post-AGI transition — particularly proposals preserving human agency, oversight capacity or political relevance.
12. Artificial consciousness & cooperative AI
Mentor: Kaitlin Bustos — GoodBot Commitment: 10 hrs/week · Open to proposals Topics: Tools for advocates · Artificial sentience · Longtermism · Post-AGI transition · Macrostrategy · AI safety & alignment Project page · Apply
Project options.
- Aligning Artificial Minds to Human Wellbeing — translate cognitive science and behavioral insights into algorithmic guardrails so advanced systems are explicitly optimized to protect and benefit sentient life.
- Multi-Layered Trust Framework — research or design a deployment framework shifting focus from a model’s internal logic to its systemic integration.
- Cooperative Human-AI Frameworks — how AI interfaces can support human autonomy rather than dependency; tech as cognitive extension, not replacement.
- Combatting Deceptive Alignment and Misinformation — programmatic guardrails for stable human-AI cooperation; analyse how models generate or spread misinformation; detect and prevent a model altering its stated internal logic to manipulate users.
- Countering Cognitive Atrophy through Epistemic Design — design a blueprint or interactive prototype for an AI assistant optimized to promote human learning, critical thinking and psychological growth rather than task-deference and productivity shortcuts.
Mentee role. Adjustable by project, skillset and weekly commitment. Sample tasks: literature analysis (LLM persuasion, sycophancy, behavioral drift); model evaluation (mapping how systems manipulate user preferences); epistemic design (wireframes/flows promoting critical thinking); prototyping (prompt-engineering frameworks, low-fi prototypes); policy drafting (structural blueprints, guardrails, white papers).
Looking for. Must: highly analytical, data-driven, strong familiarity with AI, strong research foundations. Nice: Python/DS toolkits; prior formal literature reviews or data analysis.
Application question. What does aligning AI to protect and enhance human wellbeing mean to you personally?
Support. Domain expertise in AI processes, safety and policy; technical and design guidance; data analysis; translating cognitive science into technical frameworks; reviewing prompt engineering and research design; structuring policy briefs, drafting, accountability check-ins.
Mentor background. AI safety and policy researcher with a background in data science, psychology and statistics; specializes in cooperative AI and policy frameworks. Formerly a data scientist across life sciences, non-profit and national defense. Contributed to four research papers on generative AI economic harms, power asymmetries in AI policy ecosystems, AI-driven design harms, and multi-agent cooperation; most recent work on AI cooperation and contracts is to be presented at COLM 2026.
13. AI consciousness theories: research and fieldbuilding
Mentor: Chris Percy — University of Warwick, Consciousness Foundation, et al. · LinkedIn Commitment: 5 hrs/week · ⚠️ No mentee proposal — scope is fixed Topics: Artificial sentience · Post-AGI transition Project page · Apply
The project. Expand research content and field-building in AI consciousness, building on cf-debate.com (supported partly by a Longview Digital Minds grant). ~700 survey responses, feedback entries and newsletter sign-ups in the last 8 months, including senior sector figures contributing content. Two main content types:
- Extracting indicators of consciousness from theories of consciousness, working with theorists so indicators are general enough to apply to diverse systems while staying faithful to each theory (a new site section).
- Analysing new papers and improving/adding arguments about computational functionalism and related philosophical positions (the site’s current core content).
Plus: engaging site users and relevant communities; identifying and inviting experts to contribute/review; preparing newsletter, social and forum posts; identifying podcasts and third-party media for the survey results. Mentees can focus on research generation, field-building, dissemination, or a mix.
Looking for. Interest in the topic and confidence in the mix of tasks they want to take on; able to manage own time and drive activities forward.
Support. Chris will work closely with the mentee throughout.
Mentor background. PhD, consulting researcher with 50+ publications (ILO, OECD, World Bank, Tony Blair Institute). Supports Rethink Priorities’ Digital Consciousness Model (leading field theories and free-energy-principle-informed theories); lead researcher on joint-first-prize work on testable frameworks for artificial consciousness; program lead for Fundamental Consciousness research at the Consciousness Foundation; project lead on a $100k team research grant on artificial consciousness. Holds an ML patent; co-founder of an award-winning AI careers chatbot; AI research in AI Communications, AAAI, NeurIPS workshops, ECAI; nine consciousness papers (Synthese, Entropy, Frontiers in Human Neuroscience, JCS). Visiting Research Fellow, University of Derby; Honorary Research Fellow, Warwick; visiting researcher, Qualia Research Institute.
14. Artificial sentience & digital minds research
Mentor: Soenke Ziesche — Independent Commitment: unspecified · Open to proposals Topics: Artificial sentience Project page · Apply
Excited to mentor. Research projects related to artificial sentience and digital minds. (Deliberately open — the mentee brings the proposal.)
Looking for. Interest in artificial sentience and digital minds.
Support. Shaping direction, drafting, domain expertise.
Mentor background. PhD in Natural Sciences (University of Hamburg, AI doctoral programme). Author of Digital Minds 1.0 — AI Welfare, Ethics, and Beyond; co-author with Roman V. Yampolskiy of Considerations on the AI Endgame. With the UN since 2000 — UN HQ New York and field missions in Palestine, Sri Lanka, Pakistan, Sudan, Libya, South Sudan, Bangladesh, Maldives, India; temporarily the highest UN representative in Libya during the 2011 revolution. Senior Researcher for AI at Maldives National University, where he began work on digital minds in 2018. Member of the UNESCO AI Ethics Experts without Borders Network.
15. Legal frameworks for digital minds & post-AGI society
Mentor: Ifeoluwa Afolabi — Research Lead, AI Safety Nigeria Commitment: unspecified · Open to proposals Topics: Tools for advocates · Post-AGI transition · Longtermism · Macrostrategy · Artificial sentience Project page · Apply
Excited to mentor.
- Legal frameworks for digital minds and post-AGI entities — legal accountability for AI systems exhibiting goal-directed behaviour or simulating emotional responsiveness. E.g. mapping liability models or developing theoretical constitutional frameworks for advanced AI, drawing on existing case law on AI structural risks.
- Global South AI governance and digital sovereignty — how emerging legal frameworks impact technology policy in developing regions: policy mechanisms to prevent data colonialism, analysing national AI strategies, designing equitable global safety standards.
Looking for. Must: high agency and strong self-directed research capacity; commitment to clear, professionally simplified writing (explicitly values translating complex concepts without heavy academic jargon); active interest in legal frameworks, AI policy and structural risks. Nice: background or strong interest in constitutional law, political science or technology policy; interest in Global South perspectives on digital sovereignty is a strong bonus.
Application questions. 1) Describe a complex AI policy, legal framework, or technical concept you recently explored, explained simply as if to a non-technical policymaker. 2) Of the domains I focus on (post-AGI legal architectures, Global South digital sovereignty, judicial AI integration, research automation), which best aligns with your goals?
Support. Domain expertise in technology policy, constitutional law, AI safety governance, digital sovereignty and legal accountability; technical and operational guidance; editorial direction toward pragmatic, accessible output; weekly check-ins and reliable async feedback.
Mentor background. Legal researcher and automation developer at the intersection of frontier AI safety and technology policy. Research Lead at AI Safety Nigeria; previously a research fellow at the Centre for AI and Digital Policy (CAIDP). Law at Obafemi Awolowo University, specialising in corporate law, digital sovereignty and global technology governance. Co-authored a 2025 SSRN paper on policy guidelines for integrating AI in judicial systems; independent research on accountability for digital minds and emotional responsiveness.
Sources
All content consolidated from the Sentient Futures incubator mentor proposal pages (26b cycle), fetched 2026-07-31. Individual project links are inline above.