Monastery 5-year focus research: education, gen-entertainment, roll-ups, personal medicine - #7
Open
Gerstep wants to merge 9 commits into
Open
Monastery 5-year focus research: education, gen-entertainment, roll-ups, personal medicine#7Gerstep wants to merge 9 commits into
Gerstep wants to merge 9 commits into
Conversation
…hreat, budgets, regulation Co-authored-by: stepan <gershuni.stepan@gmail.com>
…ope, continual learning, RSI Adds sourced deep-dive on AI capability slope as of 2026-09-01: - raw/agent-tech-researcher-takeoff.md: SOTA models Aug/Sep 2026, continual learning papers+products, automated AI research/RSI, world models, remaining blockers, economic layer implications - macro/rsi-continual-learning.md: GP-facing synthesis testing the capability-vs-diffusion thesis - raw/_scratch-takeoff-notes.md: raw research notes with primary source quotes Key findings: no lab shipped weight-level continual learning (both OpenAI Dreaming v3 and Claude Dreaming are memory consolidation); RSI runs at the scaffold layer with METR expenditure horizons of $0-3.3K; Anthropic reports internal R&D 'significantly faster but not yet by a factor of 2'; US firm AI adoption flat at 17-20%. Co-authored-by: stepan <gershuni.stepan@gmail.com>
…ncumbents, budget holders, Big Tech threat, 2023-26 failures Co-authored-by: stepan <gershuni.stepan@gmail.com>
…ech deep-dive Adds tech-researcher analysis for the Generative Netflix track: - video gen SOTA (Veo 3.1 / Kling 3.0 / Seedance 2.0 / open weights) with cost, audio, resolution - real-time streaming systems and measured per-GPU-second throughput - interactive world models (Genie 3, Matrix-Game 3.0, Oasis 3, AlayaWorld, Magpie, Visko Orbis) - AR vs diffusion vs hybrid architecture lineage through Flex-Forcing - cost-curve verification and crossover math against Netflix revenue-per-viewing-hour - consistency (character 0.74, Physics-IQ 39.5) as the binding constraint - lab-vs-startup defensibility and 12/24/60-month probabilistic forecast Co-authored-by: stepan <gershuni.stepan@gmail.com>
Deep market research on generative entertainment as of 2026-09-01 for the Monastery focus-areas study. Key findings: - Sora shut down Apr 2026 (1% D30 retention, $2.1M lifetime rev vs ~$1M/day cost); Disney's $1B OpenAI deal died with it, $0 realized - Generative video costs $108-1,440/hour vs Netflix's $0.084 per view-hour of amortized content: a 1,281-17,082x structural gap, since per-user generation has an amortization denominator of 1 - Every lab left consumer generative entertainment for enterprise tooling (Kling ~$500M ARR), robotics (Decart, Luma, Runway) or death - Realized value is entirely supply-side: Netflix ~300 titles + InterPositive ($587M), audiobook narration $200-500/hr -> $5-22/hr - Investable shape is 'generate the asset/state, render conventionally' Adds: - raw/agent-market-researcher-gen-entertainment.md (raw source log, 464 lines) - generative-netflix/market.md (synthesis incl. Monastery filter) Co-authored-by: stepan <gershuni.stepan@gmail.com>
… synthesis Deep research workspace for Monastery 5-year focus areas (education, generative entertainment, AI roll-ups) under the capabilities-vs-diffusion thesis. Adds company catalogs, 2026 roll-up thesis, quality review, follow-up on Higgsfield/Crosby-shaped comps, and the concise slides/report. Co-authored-by: stepan <gershuni.stepan@gmail.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
Gerstep
marked this pull request as ready for review
September 1, 2026 19:55
…companies) - raw/agent-company-researcher-pm-therapy-coaching.md: full agent output covering 46 companies across 12 subsegments (employer MH, clinician marketplaces, AI chatbots, coaching, nutrition, sleep, fertility, menopause, DTC pharma, wearable AI, ADHD aftermath) - personal-medicine/companies-therapy-coaching-consults.md: curated catalog with Monastery filter, moat analysis, regulatory snapshot (WOPR Act + 4-state AI therapy bans), and what-works vs what-failed synthesis Key findings: DTC cash-pay therapy collapsing (BetterHelp -12-19% 2026); employer B2B + insurance-panel models dominating; 4 US states ban AI therapy (IL/NV/RI/ME July 2026); licensed clinician in the loop is non-negotiable moat; Spring/Alma ~cd /workspace && git add research/monastery-focus-areas/0901-monastery-focus-areas-26/raw/agent-company-researcher-pm-therapy-coaching.md research/monastery-focus-areas/0901-monastery-focus-areas-26/personal-medicine/ && git commit -m "research: add therapy/coaching/soft consultation company catalog (46 companies) - raw/agent-company-researcher-pm-therapy-coaching.md: full agent output covering 46 companies across 12 subsegments (employer MH, clinician marketplaces, AI chatbots, coaching, nutrition, sleep, fertility, menopause, DTC pharma, wearable AI, ADHD aftermath) - personal-medicine/companies-therapy-coaching-consults.md: curated catalog with Monastery filter, moat analysis, regulatory snapshot (WOPR Act + 4-state AI therapy bans), and what-works vs what-failed synthesis Key findings: DTC cash-pay therapy collapsing (BetterHelp -12-19% 2026); employer B2B + insurance-panel models dominating; 4 US states ban AI therapy (IL/NV/RI/ME July 2026); licensed clinician in the loop is non-negotiable moat; Spring/Alma ~$1B ARR profitable; Grow Therapy $1B ARR at $3B."B ARR profitable; Grow Therapy cd /workspace && git add research/monastery-focus-areas/0901-monastery-focus-areas-26/raw/agent-company-researcher-pm-therapy-coaching.md research/monastery-focus-areas/0901-monastery-focus-areas-26/personal-medicine/ && git commit -m "research: add therapy/coaching/soft consultation company catalog (46 companies) - raw/agent-company-researcher-pm-therapy-coaching.md: full agent output covering 46 companies across 12 subsegments (employer MH, clinician marketplaces, AI chatbots, coaching, nutrition, sleep, fertility, menopause, DTC pharma, wearable AI, ADHD aftermath) - personal-medicine/companies-therapy-coaching-consults.md: curated catalog with Monastery filter, moat analysis, regulatory snapshot (WOPR Act + 4-state AI therapy bans), and what-works vs what-failed synthesis Key findings: DTC cash-pay therapy collapsing (BetterHelp -12-19% 2026); employer B2B + insurance-panel models dominating; 4 US states ban AI therapy (IL/NV/RI/ME July 2026); licensed clinician in the loop is non-negotiable moat; Spring/Alma ~$1B ARR profitable; Grow Therapy $1B ARR at $3B."B ARR at B. Co-authored-by: stepan <gershuni.stepan@gmail.com>
…ap + regulatory analysis
Add fourth Monastery focus area — personal medicine — with full investment researcher output:
raw/agent-investment-researcher-personal-medicine.md:
- Value chain map (collection → interpretation → Rx/compounding → adherence → consult)
with margin analysis at each layer
- 10-subsegment heat map (TOO LATE / OPEN / TOO EARLY):
🔴 GLP-1 telehealth, biomarker memberships, premium longevity clinics
🟡 Peptide protocols (BPC-157, MOTS-c, TB-500) — PCAC Jul 2026 advisory
🟢 AI-native DPC operator, personal physician/concierge
- Regulation as moat vs kill-switch:
- Kill-switches: GLP-1 compounding (FDA 2025); 4-state AI therapy ban (Jul 2026);
EU AI Act medical (Aug 2026); DOJ Done Global CPOM criminal theory (Jan 2026)
- Moats: 503A pharmacy + licensed prescriber; MSO-PC structure; cash-pay Stark-free
- Analogy verdict: CROSBY shape (own the licensed entity + workflow data), not
MagicSchool (wrapper killed by GPT/FDA) or Alpha School (capex kills M)
- Psychotherapy: 'soft consult' = ChatGPT feature for DTC; $1B path = payor B2B
(Talkspace $229M rev, 22% YoY — 75% payor, DTC revenue fell 30.4% Q4 2025)
- $2M/12-week build: MSO-PC structure + VITL 503A network + AI interpretation
+ adherence app + first 100 cash-pay members at $200-400/mo
- Cited: Function $450M CVF (Jul 2026), Hims/Novo deal (Mar 2026), PCAC peptide
vote (Jul 2026), state AI therapy bans, FDA generative AI mental health status
personal-medicine/thesis.md: GP-readable synthesis with heat map, regulatory
framework, analogy analysis, investment opportunities, and pass zones
slides.md: Section 6 added (personal medicine) — consistent with capabilities-vs-diffusion
frame; sections 7 (batch filter) and 8 (unknowns) renumbered accordingly;
added personal medicine watch items to section 8
Co-authored-by: stepan <gershuni.stepan@gmail.com>
…ides Add the fourth focus area (offline personalized care delivery, not pharma R&D) to the concise deck and full report. Clean duplicate slide 7 leftover, flag Function member/ARR conflicts, and treat PCAC peptide votes as advisory only. Co-authored-by: stepan <gershuni.stepan@gmail.com>
cursor
Bot
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Deep research pack for Monastery strategy (seed–A, $2M uncapped SAFE + 5%, 12 weeks).
Start here:
research/monastery-focus-areas/0901-monastery-focus-areas-26/slides.mdFull synthesis:
report.md. Evidence ineducation/,generative-netflix/,ai-rollups/,personal-medicine/,macro/, plus raw agent outputs.Headline for Batch 2 sourcing:
23 agents + quality review. Research MCPs were unavailable; sourcing is web search/fetch with citations. Extraordinary private prints (Function member counts, Neko ~$7B, VITL ARR, Higgsfield $700M) are flagged as company/media/self-reported.