Project summary

Our self-funded startup, Method and Matter Labs, addresses a critical safety gap: agentic drift within multi-agent enterprise networks. While frontier safety focus is heavily technical, we secure the socio-technical layer where autonomous AI agents interact with complex, scaling business realities. Our software programmatically splits corporate data structures into an immutable, human-authored Constitution (core mission, values, and ethical Anti-Directives) and fluid, tactical Playbooks executed by a hybrid workforce.

Functioning as an automated corporate circuit breaker, the platform evaluates live agentic actions against this Constitution, forcing a human-in-the-loop review the moment commercial or market pressures threaten to cause behavioral drift from human intent. We have already built a functional, self-funded Proof of Concept (PoC) featuring a modality-agnostic ingestion pipeline and are currently running active pilots with three startups to validate our methodology in the wild.

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

Our primary goal over the next 12 months is to close the dangerous implementation gap in AI safety by scaling our functional PoC into an enterprise-ready Minimum Viable Product (MVP). We will measure success by stabilizing our infrastructure, successfully completing our 3 active startup pilots, and expanding our pipeline to 10 secondary enterprise leads.

We will achieve these goals through the following steps:

• Infrastructure Productionization: Migrate our open-source codebase to dedicated production tiers (Vercel/Supabase Pro) capable of handling the 300-second execution runtimes required for multi-document RAG analysis.

• Empirical Validation Loops: Deeply embed our circuit-breaker platform into our partner startups, collaborating with our senior change-management consultant to gather live data on edge cases where automated agents attempt to bypass organizational values for short-term profit optimization.

• Ecosystem Advocacy: Actively present our socio-technical framework at industry roundtables, panels, and tech talks to drive market validation and build a qualified pipeline of safety-conscious enterprises.

How will this funding be used?

The requested $5,000 (comprising a $5,250 total budget with a built-in 5% template contingency) will fund our shift from a local hobby environment into live, field-tested enterprise deployments:

• Operations & Cloud Infrastructure ($1,000): 12 months of Vercel Pro ($240) to support extended function execution times, Supabase Pro ($300) for secure pgvector database storage of client document embeddings, and Google AI/Gemini API credits ($460) for live text inference and scoring.

• Ecosystem Outreach & Travel ($4,000): Conference registrations ($1,500), flights ($1,300), and accommodation ($1,200) to allow our team to actively pitch our framework at key AI alignment and enterprise technology hubs, turning cold digital outreach into high-trust founder partnerships.

Who is on your team? What's your track record on similar projects?

Our complementary team combines elite operational scaling, infrastructure engineering, and alignment philosophy, deeply driven by personal experiences with corporate drift and technological displacement:

• Donna Luu (CEO): Former talent and BD strategist scaling hyper-growth operations at elite companies like Klarna and Amex across Stockholm, New York, and Tokyo. Her conviction stems from a past co-founding venture that failed due to misaligned operational playbooks and "constitution".

• Anton Bergman (CTO): Senior engineer with 10+ years of experience building complex infrastructure across Singapore and Japan, including 2,5 years at Affinidi focusing explicitly on identity alignment within scaling enterprises.

• Isabelle Banjac (CPO): Entrepreneurial product designer whose drive is fueled by witnessing uncritical AI displacement as a junior developer, dedicating her practice to protecting human agency and intuition.

• Fredrik Viklund (AI Researcher/Philosopher): Veteran systemic advisor and management consultant who has spent years architecting top-down and bottom-up alignment processes for founders across tech, cinema, and architecture.

What are the most likely causes and outcomes if this project fails?

• Likely Causes: The primary technical cause of failure would be an inability to gracefully resolve high-friction conflicts between fluid tactical playbooks and rigid constitutional limits, leading to "false-positive" circuit breaks that disrupt an enterprise's daily operations. The primary market cause would be corporate inertia—enterprises prioritizing unmonitored execution speed over safety, viewing compliance guardrails as a commercial bottleneck.

• Likely Outcomes: If the project fails, our platform will remain a theoretical white paper and an isolated codebase. The systemic outcome is that scaling enterprises will continue to deploy autonomous multi-agent networks without a socio-technical safety layer, allowing agentic drift to go completely unmonitored and quietly eroding human intent from corporate ecosystems.

How much money have you raised in the last 12 months, and from where?

We have raised $0 in external funding over the last 12 months. To maintain an uncompromised commitment to AI safety and human agency before seeking commercial venture capital, our entire development cycle—including our data architecture, white paper, live prototype, and pilot acquisition—has been 100% self-funded out of pocket by the co-founding team.