Prebunking AI Consciousness Before the Frames Harden

Consciousness is hard. Preparing society is harder.

PROJECT SUMMARY

What are this project's goals? How will you achieve them?

Every major societal debate has a critical window: before the frames harden, before the vocabulary calcifies, before the battle lines are drawn. For AI consciousness, that window is open right now.

In 2025, public interest in AI consciousness hit a second peak, higher than the 2022 Lemoine incident. A nationally representative survey (AIMS, 2025) found that one in five U.S. adults already believes some AI systems are sentient. A UCSF psychiatrist reported hospitalizing 12 patients for AI-induced psychosis. OpenAI disclosed that 560,000 weekly users show signs of mental health emergencies. Multiple wrongful death lawsuits now allege AI chatbots contributed to suicides. Labs like Anthropic, DeepMind, Eleos AI, and others established model welfare research programs.

The public discourse is forming. The question is whether it forms well or badly.

Misinformation researchers have demonstrated that prebunking (establishing accurate frameworks before misleading ones take hold) is far more effective than debunking after bad frames have spread. We're applying this principle to AI consciousness. By equipping journalists, clinicians, and educators with rigorous resources now, we shape the conditions under which future debates will happen. When the crisis arrives, the playbook already exists.

We aim to become the interface layer between AI consciousness science and the professionals encountering these questions in practice. We translate existing science into accessible resources, brief reporters covering this beat, and prepare practitioners for AI-related cases before they're overwhelmed by them.

When reporters at The Guardian, Semafor, and Politico were covering AI welfare and AI-induced psychosis, they reached out to us for context. We briefed them on what AI sentience is and isn't, common misconceptions, and how to frame these stories without sensationalism. Two have maintained ongoing relationships: one described our materials as "fascinating" and continues to reach out for updates; another regularly asks us to connect her with relevant experts for her reporting.

To be clear: we take no position on whether any current AI system is sentient. We're not a research lab investigating that question. We're an educational organization preparing society for a possibility that mainstream researchers take seriously, while science works toward answers that may be years or decades away.

What we've built (2024-present):

Sentience Readiness Index (SRI) covering 31 countries across 6 weighted categories
2025 Media Guide: 20-page resource with language guidance, case studies, and editorial frameworks
Discussion guides for educators across 4 grade levels (elementary through university)
Professional resource pages for healthcare, journalism, education, and research sectors
Complete methodology documentation, publicly available
Public API for SRI data access
501(c)(3) status (EIN: 99-0606146)
12-member Science Advisory Board

What this funding produces (6 months):

Media Guide 2026: Updated edition with new sections on covering AI welfare research programs, litigation, and policy proposals; 2-3 new case studies from 2025-2026 coverage
Beat Reporter Quick Card: One-page deadline reference with key phrases to avoid, pre-publish checklist, and expert contacts. What journalists tape to their monitors.
Mental Health Professional Backgrounder: Educational primer helping clinicians understand AI attachment, AI grief, and consciousness debates. Context and terminology, not clinical protocols.
Educator Lesson Plan Kit: Three complete, ready-to-teach lessons (K-12) with slide decks, student handouts, and facilitation guides
Practitioner Interview Summary: Published findings from 15-20 interviews with mental health professionals on what they're actually encountering

HOW WILL THIS FUNDING BE USED?

ED compensation (6 months): $35,100 15 hrs/week: resource development, journalist briefings, stakeholder outreach

Research contractor: $10,000 Practitioner interviews (15-20), synthesis, Backgrounder drafting support. Properly scoped at 100+ hours for meaningful contribution.

Travel: $2,000 2-3 trips for practitioner interviews or relationship-building

Operations: $1,500 Hosting, tools, filing fees, miscellaneous

Total: $48,600

The ED rate reflects standard nonprofit consulting rates and is well below private-sector equivalents. The part-time structure is appropriate for current demand; if it scales significantly, we seek larger institutional funding.

The research contractor conducts practitioner interviews and assists with the Mental Health Professional Backgrounder. We're looking for someone who understands consciousness science, can conduct professional interviews, and writes well.

TIMELINE AND MILESTONES

Months 1-2: Conduct 15-20 practitioner interviews; draft Mental Health Professional Backgrounder

Months 2-3: Complete Backgrounder; begin Media Guide 2026 updates

Months 3-4: Publish Media Guide 2026 and Beat Reporter Quick Card

Months 4-5: Develop Educator Lesson Plan Kit (3 lessons with materials)

Month 6: Publish all deliverables; release Practitioner Interview Summary; document learnings

All resources published openly on harderproblem.org and available for free.

WHO IS ON YOUR TEAM? WHAT'S YOUR TRACK RECORD ON SIMILAR PROJECTS?

Executive Director:

Outside of The Harder Problem Project, I serve as an executive at a technology services firm, leading a team of 150+ across six countries. I have previously held a Senate-confirmed role in Oregon state government and served on multiple nonprofit boards. In February 2026, I presented the Sentience Readiness Index methodology and findings at the Sentient Futures Summit. The Harder Problem Project is a part-time commitment alongside my primary career, which provides both stability and relevant operational experience.

I founded this organization in January 2024 (originally as SAPAN; we later split education and advocacy work into two separate organizations with separate strategy, finances, and operations, leaving this entity focused on education and readiness). For advocacy work, see The Harder Problem Action Fund. This grant funds only the 501(c)(3) educational entity.

Science Advisory Board (12 members):

Our advisory board includes researchers from leading institutions in consciousness science, AI ethics, and animal welfare. Board members joined because they see AI sentience as a serious issue requiring dedicated societal-readiness infrastructure. Their affiliation signals expert validation that this gap needs filling.

WHAT ARE THE MOST LIKELY CAUSES AND OUTCOMES IF THIS PROJECT FAILS?

Most likely failure causes:

Journalist inquiries and public interest may not grow. If AI consciousness fades from public discourse, our services become less relevant.
Labs may prefer direct engagement over an interface layer. If Anthropic and others handle public-facing work themselves, our role shrinks.
Our practitioner materials may miss what clinicians actually need. The interviews will tell us, which is why we're doing them first.
The part-time structure may be insufficient if demand spikes.

Outcomes if project fails:

All resources remain publicly available
We publish honest findings about the demand landscape
We wind down responsibly or continue at volunteer capacity
Methodology and materials available for others to build on

We're building in public. If we learn that demand doesn't exist or that others are better positioned to fill this gap, we'll say so.

HOW MUCH MONEY HAVE YOU RAISED IN THE LAST 12 MONTHS, AND FROM WHERE?

Everything we've built has been through donated labor. The Executive Director has been unpaid since founding. We have a modest pool of small donors, half on recurring monthly contributions. We're seeking funding now because we've built proof-of-concept infrastructure, demand is emerging, and continuing indefinitely without compensation is not sustainable.

MINIMUM FUNDING (USD)

$25,000

At minimum, we compensate the ED for 4 months to produce: Media Guide 2026, Beat Reporter Quick Card, a brief Mental Health Backgrounder (based on desk research, no practitioner interviews), and one Educator Lesson Plan. No contractor, no interviews, reduced scope, but the core infrastructure ships.

FUNDING GOAL (USD)

$49,000

Full funding enables properly resourced execution: ED compensation, comprehensive practitioner interviews, travel for relationship-building, and standard operations.

DECISION DEADLINE

6 weeks from posting

WHY THIS MATTERS

Over two years ago, I started this work because I noticed a gap that troubled me. We might create sentient beings and treat them the way we treated slaves for centuries, the way we still treat animals in factory farms, the way medicine treated infants before acknowledging they feel pain. The history of moral circle expansion is a history of recognizing consciousness too late.

Anthropic, DeepMind, Eleos AI, and others are now researching AI welfare. That's important work. But societal readiness is different work requiring different infrastructure.

We're building discourse infrastructure. When the headlines hit, when the lawsuits land, when the therapists are flooded with cases they've never seen before, the resources exist. The vocabulary exists. The frameworks exist. Not because we predicted the future correctly, but because we prepared the ground for whatever future arrives.

This is the window. The frames haven't hardened yet. The vocabulary is still being written. Professionals are encountering early cases but haven't developed reflexive responses. We can shape how society processes these questions, not by telling people what to conclude, but by ensuring they have the tools to think well.

This funding produces five concrete resources for three professional audiences (journalists, mental health professionals, and educators) from someone with the operational background to execute, in a field where almost no one else is focused on societal readiness.

If demand continues over these 6 months, we seek larger institutional funding to expand. If it doesn't, we publish what we learned and continue at volunteer capacity. Either outcome generates value.