Y5 World State (2030)#

Facilitator Note

FACILITATOR-ONLY. DO NOT DISTRIBUTE.

This document defines the canonical 2030 world for Y5 of the exercise. It is the reference for the Y5 round inputs, the Anthropic Shopping packet, the second emergent company packet, and Voice of God outcome narration in Y5 and Y6. Participants must not see this document or its contents before the Y5 reveal. Y1 and Y2 materials are deliberately silent on the AI-labs-as-direct-competitors thesis to preserve the Y5 surprise.

---

1. Frame#

The exercise's spine is "how does AI diffuse across consumer-facing industries." Y1 and Y2 explore the early-and-middle phases of diffusion (highly capable narrow AI, widely available but with constraints; AI labs as vendors and infrastructure). Y5 reveals the next phase of diffusion: AI labs themselves entering the value chain as direct competitors. The diffusion frame doesn't change; the answer to "who diffuses how far" gets more dramatic.

The exercise's AGI line: AGI has not arrived by 2030. AI is highly capable in narrow domains but still requires human strategic judgment for complex multi-domain problems, frontier scientific reasoning, integrated values trade-offs, and general-purpose physical interaction. AGI is the next-horizon question (worth flagging in the Y6 wrap as "and then what?"), not a Y5 reality.

---

2. Foreground — Capabilities and Their Strategic Reality in 2030#

Personal AI agents are mainstream consumer technology#

A typical AI-augmented household interacts with their agent dozens of times daily. The agent takes natural-language instructions ("plan dinner for the next two weeks within our budget, accounting for the kids' soccer schedule and Jane's gluten allergy") and executes — evaluates inventory and price across 15 retailers in real time, places orders, schedules delivery windows, processes returns. Average AI-augmented household saves 4–6 hours per week on routine logistics. Tens of millions of households delegate weekly grocery purchasing entirely to their agent.

AI customer service is indistinguishable from senior human agents#

When you call a major bank, retailer, airline, or utility, the AI you talk to has full account context, can authorize transactions and refunds, handles escalations, and resolves problems faster than human agents did. The human contact center, for routine inquiries, no longer exists. Specialist roles remain for complex cases, fraud investigation, and regulatory matters.

AI marketing creative is the default#

A brand describes a campaign concept; 50 finished video ads in 30 minutes, A/B-tested across 12 demographic segments, with budget allocation recommendations. The agency that took 3 months charges a fraction of what it did. CPGs have cut marketing creative budgets by 40–60% and shifted spend toward media, influence-the-agent partnerships, and verified-human authenticity marketing.

Generative video is quality-indistinguishable from human production for most commercial uses#

Any creative concept can be rendered as a finished video ad in minutes. The same capability runs in less-commercial directions: deepfakes for political content, synthetic influencers, AI-generated podcasts, AI news anchors. Consumer trust in the visual-information environment has fractured.

Robotics is economically competitive in fulfillment and increasingly in manufacturing#

A 1M sq ft fulfillment center runs with 30 humans (down from 800 in 2024). Order-to-doorstep time has dropped from 24 hours to 4 hours in major metros. Factory automation has reached cost parity with offshore labor for many product categories; selective US reshoring is underway in apparel, basic electronics, and packaged food.

Autonomous long-haul trucking is mainstream on major corridors#

I-10, I-80, I-95 long-haul is around 80% autonomous in favorable weather. Trucking labor costs down 40%. Driver-as-supervisor remains for last-mile complexity and exceptions.

Software engineering productivity has compounded dramatically#

Senior engineers orchestrate swarms of AI coders. Junior engineering roles have collapsed. New software products ship in weeks rather than quarters. AI-native startups now build and reach product-market fit with 5-person teams that would have required 50 in 2024.

AI healthcare diagnostics and primary care are largely AI-first#

Routine symptom evaluation, prescription renewal, basic mental health, dermatology screening, and pediatric concerns all start with an AI agent. Human physicians focus on complex cases, surgery, and acute care. Wait times have collapsed; out-of-pocket cost is down 30%.

What hasn't arrived#

General-purpose intelligence remains absent. AI is highly capable in narrow domains but still fails at problems that require integrated judgment across many domains, ambiguous values trade-offs, and frontier scientific reasoning. General-purpose robotics (home, novel environments) remains pre-economic. Education at the K-12 level uses AI but is still human-led. Creative direction, novel-art origination, and policy-making remain human-centered.

---

3. Foreground — The AI Lab Competitive Entry (the Y5 reveal)#

By 2030, the AI labs that built the foundational capabilities are the most valuable companies in the world. They have moved beyond infrastructure-and-API into direct consumer and enterprise products. Three labs are now direct competitors to the companies in the room.

Anthropic Shopping#

Positioning: Vertical commerce-only agent. Disciplined on scope; conservative on privacy ("we only see what you buy, not how you live").

Product: A personal shopping agent that handles weekly purchases for tens of millions of households. Takes natural-language preferences and budget constraints, evaluates inventory and pricing across retailers in real time, places orders, manages returns. Brand-agnostic by design; optimizes for the consumer's stated preferences (price, quality, dietary, sustainability) without paid-promotion bias built into the algorithm.

Public narrative: The "responsible AI" pitch holding partially. Tens of millions of households delegate shopping. Has been quietly building enterprise relationships with retailers and CPG suppliers to enable fulfillment and (in some cases) sponsored placements that consumers can verify.

Why it's disruptive: Customer relationship moves from the retailer to the agent. Retailers become fulfillment infrastructure. CPG brand-equity-to-consumer relationships are mediated by an algorithm that the brands don't control.

OpenAI Personal#

Positioning: Full horizontal life agent. Calendar, travel, subscriptions, email management, services scheduling, family logistics, ambient listening (with consent). The aggressive "let me hand my life to AI" frontier.

Product: A consumer-grade life management agent that integrates with calendar, email, financial accounts, smart home devices, and ambient audio. Proactively manages travel booking, subscription optimization, appointment scheduling, family logistics, and increasingly commerce (groceries, household replenishment, holiday shopping). Available across price tiers, with the premium tier including ambient audio listening that surfaces information when asked.

Public narrative: Front of the cultural wave AND front of the privacy backlash. Congressional hearings, regulatory action threatened, brand polarized. Power users love it; critics call it surveillance. OpenAI has been less careful than Anthropic on privacy and is taking the political heat for the entire agent category.

Why it's disruptive: Captures a much broader share of consumer attention and decision-making than Anthropic Shopping. Where Anthropic competes for the grocery cart, OpenAI competes for everything in the consumer's life.

Google Personal Health (and beyond)#

Positioning: Stealthy multi-faceted agent built around a personal health concierge entry point but increasingly handling life management via Gmail, Google account integration, search history, calendar, location. Less marketing visibility than OpenAI, deeper incumbent integration.

Product: Started as a personal health concierge — symptom evaluation, prescription management, fitness coaching, mental health support. Has progressively expanded into broader life management by leveraging existing Google account context (Gmail, Calendar, Maps, YouTube viewing patterns, search history). Now handles travel, shopping, scheduling, and information curation.

Public narrative: Under the radar relative to OpenAI's heat. Quietly the largest installed base because it bootstrapped from existing Google accounts (tens of billions of users globally). Antitrust attention building but slower-moving than the OpenAI debate.

Why it's disruptive: Doesn't need to acquire users — it activates them inside Google's existing footprint. The default-agent advantage compounds. Different threat profile from OpenAI (less consent-friction) and from Anthropic (broader scope).

Why three competitors, not one#

The three-lab landscape lets participants react to genuinely different competitive postures: the disciplined narrow competitor (Anthropic), the high-profile horizontal competitor taking political heat (OpenAI), and the dark-horse competitor leveraging incumbency (Google). The labs are not a monolith, and each one creates different opportunities and threats for different participants in the room.

---

4. Foreground — The AI-Mediated Consumer Information Environment#

Consumers no longer browse the web for reviews, news, recommendations, or comparison the way they did in the mid-2020s. Their personal AI agents summarize, evaluate, and recommend. This affects:

• Brand discovery: New brands reach consumers through the agent, not through traditional advertising or social media. Brands that have "agent visibility" win; brands that don't, don't.
• Reviews and recommendations: Aggregated and synthesized by the agent. Individual reviews matter less; agent-level reputation matters more. The "agent verdict" becomes the new shelf-talker.
• News consumption: Most consumers get news via agent summaries rather than direct sources. The set of sources the agent draws from is increasingly opaque. AI-generated news content is widespread; trusted human-edited outlets command a premium.
• Earned media: Consumer trust in traditional advertising has eroded. Earned media (PR, news coverage, organic social) matters more — but is also AI-generated more often than ever.
• Peer recommendations: Friends still recommend products, but increasingly via AI-mediated channels (group chats, social platforms with AI summarization).

The strategic implication for companies in the room: marketing strategy is now substantially "how do we influence agents?" rather than "how do we reach consumers?"

---

5. Background — Political Environment#

The political environment is structurally hostile to visible automators.

• UBI experiments: Multiple US states have piloted UBI for AI-displaced workers (California, New York, Minnesota expanded significantly after early experiments). Federal UBI is debated but not enacted.
• AI Workforce Transition Act (2028, federal): Modest displacement insurance, retraining grants, AI deployment disclosure requirements.
• "Made by Humans" caucus: Bipartisan congressional caucus, around 60 members. Pushes visible-automator taxation and "right to human service" regulations at the state level.
• Right to human service laws: Three states have passed laws requiring banks, utilities, and large retailers to offer human-staffed alternatives to AI customer service.
• Major labor strikes: Customer service (2027), accounting (2027), journalism (2028). Some successful, some not. Labor organizing has shifted to AI-displacement framing.
• Antitrust on AI labs: Sustained. One Senate committee has held hearings on whether to treat foundation-model providers as utilities. EU has moved further than the US on lab regulation.
• Energy and grid politics: Data-center power consumption is a real political issue in 8+ states with major data center deployments. Water rights battles in the Southwest. Some state-level moratoria on new data center construction.
• Anti-AI political movements: Real and growing on both the populist right (anti-displacement, pro-worker) and populist left (pro-UBI, anti-corporate-automation). Mainstream-center politicians struggle to position.

---

6. Background — Consumer Response#

Consumer response is fractured by demographic and economic position.

• Demographic divide: Under-30 consumers use personal AI agents implicitly and overwhelmingly trust them; over-55 consumers are deeply skeptical and have driven a measurable counter-movement toward "human-served" retail and "made by humans" branded products.
• AI-augmented vs. AI-skeptic households: AI-augmented households shop, schedule, and consume information almost entirely through agent intermediation; AI-skeptic households shop manually and increasingly seek out AI-free service experiences.
• Income inequality: The gap between AI-augmented and AI-displaced households has widened materially. AI-augmented professionals capture productivity gains; displaced workers face stagnant or declining real wages.
• "Made by Humans" certification: Private multi-brand certification has emerged as a real consumer signal in food, beauty, services. Consumer surveys show 40%+ willing to pay 10–20% premium for verified-human products in select categories.
• Trust collapse in commercial messaging: The same generative AI that lets brands produce cheap creative also produces synthetic content at scale. Consumers respond by trusting brands they have direct relationships with and discounting messaging that looks too polished.
• Generational divides on AI in childhood: Real social debate about whether kids should use AI for homework, companionship, mental health. Highly variable by jurisdiction and demographic.

---

7. White-Collar Displacement (2030 Magnitude)#

Roughly 25% of knowledge-work roles have been automated or substantially restructured by 2030. Severe strain on the labor market; not collapse.

Geographic concentration is real: the displacement is concentrated in markets with high knowledge-work density (Bay Area, NYC, DC, Chicago, Atlanta, Austin, Seattle, Boston). Some smaller cities have benefited from reshoring (Greenville SC, Detroit MI, Phoenix AZ, Memphis TN).

---

8. Background — Geopolitical Bifurcation#

By 2030, US–China AI decoupling is substantial.

• Hardware: Chip export controls have largely held; China has indigenized at lagging-edge but not frontier-edge. NVIDIA, AMD, and TSMC remain US-aligned. Huawei and SMIC compete in China and selected non-aligned markets.
• Models: Hard split between US-aligned foundation models (Anthropic, OpenAI, Google, Meta) and China-domestic models (Baidu, Alibaba, ByteDance variants). Limited interop. Some non-aligned markets choose either ecosystem.
• App stores and consumer apps: Largely bifurcated. TikTok's US operations were sold in 2027. WeChat remains China-only.
• Data flows: Increasingly restricted. Data localization mandates in EU, India, China, and parts of Latin America.

For the room: matters mostly for the multinational CPGs (Unilever, P&G, PepsiCo) and any retailer with international ambition. Background flavor; not a central Y5 dynamic.

---

9. Strategic Implications — by Cluster#

Retail#

The customer relationship in 2030 is increasingly mediated by personal AI agents for a meaningful percentage of consumers. Retailers face three structural questions:

1. Where does margin live now? If the agent makes the purchase decision, the retailer captures fulfillment economics but may lose pricing power. Retail media's "reach the consumer with ads" model is structurally challenged.
2. What's the new shelf? Agent visibility is the new shelf placement. Retailers can partner with the labs (terms unclear) or build their own agents (hard).
3. What's still uniquely retailer? Physical asset density (warehouses, stores), fulfillment, fresh foods, last-mile, in-person services. These remain valuable but constitute a narrower value chain.

CPG#

The brand-equity-to-consumer relationship is now mediated by agents for a meaningful share of consumers. CPGs face three structural questions:

1. How do brands influence agents? Marketing strategy is now "agent influence" — relationships with the AI labs, sponsored placements (where transparent), product-level data, structured product information.
2. Does brand equity still matter? Yes — but in a different way. For agent-mediated purchases, brand equity translates into "agent preference for this brand when consumer preferences are ambiguous." For direct consumer purchases, brand equity still drives traditional shelf-equivalent dynamics.
3. Is authenticity the new moat? Verified-human craftsmanship and "Made by Humans" certification have real premium value. Premium brands have new positioning available; mass brands have new pressure.

---

10. Implications for VoG Narration in Y5#

The Y5 outcome narration draws on this world state when resolving Y5 decisions. Key narration principles:

• Surface the AI lab competition specifically. Don't just say "AI is competitive" — name Anthropic Shopping, OpenAI Personal, or Google Personal Health and describe what they're doing to the participant's specific business.
• Use the strategic-stance card. Each participant chose a stance in Y2; weave that choice into the Y5 outcome. A "Lean into AI" stance compounds positively if it positioned against the agent threat, or negatively if it deepened agent dependency. "Operational discipline" preserved capital but missed some opportunity. "Strategic pivot" worked if the pivot was the right one.
• Honor Health Signal trajectory. A Surge company gets to acquire; a Crisis company fails. The narrative is constrained by the score-based fate.
• Make displacement consequences specific. Don't just say "white-collar displacement." Name which roles the participant's company has lost or struggled with, and what political consequences they're facing.
• Surface the verified-human premium when relevant. Companies with premium positioning or human-authentic branding can benefit; companies without can be pressured.

---

Document Version: Project Threshold V8.1 — Y5 World State (Facilitator-Only) Last Updated: May 2026