Theseus Trajectory: Part B

---

“If Helion Energy's publicly announced timeline holds, the company will target a grid-connected fusion power demo in 2028 — not the 2030s. Whether or not this specific date holds, the broader pattern it represents — private fusion moving faster than state-led programmes — is already visible in 2026.”

Executive Summary: The State of the Ship (A Simulated Checkpoint)

The Ship Is Still Sailing, in This Scenario

Under the trends and announced plans visible as of 2026, the Theseus Trajectory would remain the correct framework at a 2028 checkpoint. The core thesis — that the global economy is being rebuilt plank by plank while still operational — is not something this exercise finds reason to abandon.

But Some Planks Would Likely Rot Faster Than Part A Expected

Three developments, visible in 2026 trends and announced plans, would plausibly dominate a 2028 checkpoint if they continue on their current course:

1. Fusion may not be delayed so much as diverging. Part A's original 2030 projection did not distinguish between state-led (ITER) and private (Helion, CFS) timelines. If current plans hold, Helion's grid-connected demo would target 2028¹; ITER's first plasma is currently projected for 2038–2040¹³. If both timelines hold roughly as announced, the fusion future would not be uniform — it would be bifurcated.

2. The governance gap looks likely to become the binding constraint. Industry commentary already frames the next few years as decisive for AI standards: decisions made now may shape which countries control the standards, safety frameworks, and deployment norms for the world's most advanced AI systems¹². If that pattern holds, the window for a coordinated Golden Path would be closing — and closing quickly.

3. Early Blessings Equity pilots would plausibly emerge somewhere other than expected. If sovereign wealth funds in the Gulf States and Singapore continue positioning for a post-hydrocarbon era at the pace already visible in 2026¹⁰, they — not the US, EU, or China — would be the most likely first movers. This is a pattern Part A did not anticipate.

The Key Insight: Part A's “Fragmented Transition” scenario (50% probability) would plausibly remain the modal outcome at this checkpoint — but split into two distinct sub-scenarios, if current bloc-level dynamics continue.

1.0 · What This Scenario Suggests We Would Get Right

The remainder of this report proceeds as a simulated 2028 checkpoint. Everything below describes what the Theseus framework would plausibly show if 2026's visible trends and announced plans continue roughly as they stand — not a record of events that have occurred.

1.1 The Theseus Framework Would Likely Be Vindicated

The central metaphor — rebuilding the ship while it is still sailing — is already gaining traction with unexpected stakeholders as of 2026, and this scenario assumes that trend continues: growing adoption by central bankers and multilateral officials in the framing of their own long-horizon communications.

1.2 The Six Filters Would Likely Remain the Binding Constraint

Part A identified the Six Filters as the human-made obstacles to the $1Q transition (see Part A for the original framework)¹⁵. Extrapolating from evidence already visible in 2026, this scenario assumes technological progress is not the bottleneck — governance and distribution are.

1.3 The Growth Model Would Likely Prove Directionally Correct

The compounding logic of Part A appears sound so far. Extrapolating 2026's visible growth signals to a 2028 checkpoint, global GDP growth would plausibly track closer to the 12% Conservative case than the 15% Base Case. The underlying engines — AI, emerging markets, connectivity — would be firing roughly as Part A predicted.

Illustrative 2028 GDP under this scenario: ~$118T

Part A's original 2028 projection (15% Base Case): ~$139T

Under this scenario, the shortfall would be almost entirely explained by the fusion divergence and governance friction described below — not by a failure of the underlying growth engines.

2.0 · Where This Scenario Suggests the Original Framework Would Need Revision

2.1 Fusion May Not Be Delayed So Much As Diverging

This is the most significant error. Part A assumed a single, unified fusion timeline. The 2028 reality is more complex:

The Implication: the fusion future is not uniform. It is bifurcated between a faster, riskier private track and a slower, more methodical state track. The “Energy Savings” engine (+3% per year) may fire earlier than Part A's conservative revision suggests — but only for jurisdictions that back the private track.

2.2 Blessings Equity Adoption Path

Part A assumed Blessings Equity would be adopted first by resource-importing, population-heavy nations. The 2028 reality is different.

First Pilots

Singapore — Launched a “Sovereign AI Dividend” in 2027, funded by a 2% levy on AI infrastructure revenue generated in the city-state.

UAE (Dubai) — Announced a “Future Wealth Fund” in early 2028, capitalised by a portion of its sovereign wealth fund returns.

Norway — Expanded its existing sovereign wealth fund model to include AI and fusion revenues.

Why These First? These are resource-exporting or hub economies with existing sovereign wealth infrastructure. The adoption path is not “global coordination followed by national implementation” — it is “early adopters implement piecemeal, creating pressure on laggards to follow.”

2.3 AI Productivity Concentration

Part A projected AI & Automation as a +5% global engine. The 2028 reality reveals a more nuanced picture:

US: AI productivity gains are visible in software, logistics, and professional services. The US maintains a 12–24 month lead over China — but this lead is narrowing.

China: DeepSeek models are priced “dozens of times” cheaper than Anthropic's offerings⁸, giving China a competitive advantage in global markets. China's strength in real-world AI applications could “shift the balance of power” by 2028².

EU: Regulatory constraints (AI Act) have slowed deployment, but the “quality” of deployment is higher.

India & ASEAN: AI adoption is accelerating in services (IT, BPO, fintech), but infrastructure gaps remain.

The Cognitive Divergence Reality: recent research reveals a “Cognitive Divergence Theory” — productivity gains from AI are not evenly distributed but proportional, often exponentially, to the user's cognitive capacity. Within 24 months of adoption, the labour market undergoes not gradual stratification but categorical divergence⁴.

The Global +5% Engine Is Real, But Uneven. The concentration of gains is exacerbating the inequality trap.

2.4 The Workforce Paradox Was Underestimated

Part A framed AI as a workforce displacer. The 2028 reality reveals a more complex picture:

Overcapacity in Legacy Roles: 92% of C-suite executives report up to 20% workforce overcapacity due to automation. By 2028, 40% of leaders expect 30–39% excess capacity⁹.

Acute AI Talent Shortages: 94% of leaders face AI-critical skill shortages today, with one in three reporting gaps of 40% or more; 44% still anticipate gaps of 20–40% in critical roles by 2028⁹.

The Demographic Fill: AI and robotics are not just replacing workers — they are filling a demographic hole created by shrinking working-age populations. The working-age population in the Global North is shrinking fast, and without AI and robotics, GDP growth would be headed for structural decline.

The Implication: the “AI replacement” narrative is too simple. The reality is a dual challenge — workforce redundancy in legacy roles and AI skill scarcity in critical roles. The Theseus transition is not just about replacing old planks with new ones; it is about managing the paradoxical coexistence of overcapacity and scarcity.

3.0 · The Revised Scenario Fan

Based on 2026–2028 data, the probabilities and outcomes have shifted.

3.1 The Four Scenarios (Updated 2028)

3.2 What This Means

The “Fragmented Transition” Is Now the Central Case. The original advice — “plan for Fragmented Transition, hope for Golden Path” — is now the primary strategic posture.

The Governance Window Is Closing: Anthropic's 2028 paper frames the coming two years as decisive. Decisions made in 2026–2028 will shape which countries control the standards, safety frameworks, and deployment norms for the world's most advanced AI systems²².

4.0 · The Filters Revisited: Which Are Binding?

4.1 Filter 1: Distribution — Now Binding

What This Scenario Assumes

AI-driven productivity gains have accrued disproportionately to capital owners.

The wealth gap has widened, with the top 1% now holding ~47% of global wealth²⁶.

Why It Would Be Binding, If This Path Holds

“Blessings Equity” pilots are happening, but they are small-scale and not yet universal.

The political backlash is growing.

4.2 Filter 2: Governance — Now the Most Binding Filter

What This Scenario Assumes

US and China have diverged on AI standards. The US model is “light-touch” with some federal oversight; China's model is “state-led” with centralised control.

The EU has adopted the world's first comprehensive AI Act, but it is not aligned with either US or Chinese standards.

Why It Would Be Binding, If This Path Holds

Multinational companies face compliance costs that are 2–3x higher than they would be with a single framework.

The “regulatory arbitrage” dynamic is accelerating.

What Would Need to Happen

Focus on interoperability, not unification.

Establish a “Basel Committee for AI” — a soft-law body that sets benchmarks and reporting standards¹⁷.

4.3 Filter 3: The Great Repricing — Diverging, Not Delayed

What This Scenario Assumes

Fossil fuel prices have remained volatile but not collapsed.

“Fusion bonds” have not materialised because the fusion timeline is uncertain.

Why It's Not Binding Yet

The fusion divergence has given the market time to adjust.

What Would Need to Happen

Prepare for the repricing to happen in different tracks — faster in jurisdictions backing private fusion, slower in those backing state-led.

4.4 Filter 4: Cognitive Bottleneck — Becoming Binding

What This Scenario Assumes

The volume of AI-generated information has exceeded human processing capacity in most large organisations.

“Cognitive Divergence” is real — the gap between AI-symbiotic workers and displaced workers is widening²⁴.

Why It Would Be Binding, If This Path Holds

The “Data Famine” is here.

The “School Without Walls” has not been built.

4.5 Filter 5: Human Factor — Moderately Binding

Anti-AI sentiment has grown, but it is not yet a majority view.

Populist parties in Europe and North America are using AI as a wedge issue.

4.6 Filter 6: Ethical/Environmental — Emerging as Binding

AI training energy consumption has created new water-stress issues.

Data centres in water-scarce regions are competing with local communities for water.

5.0 · The Blessings Equity Pilot: A Case Study

5.1 The Singapore “Sovereign AI Dividend”

What Has Worked

Simple administration via existing tax and social transfer infrastructure.

Broad political support.

What Has Not Worked

The amount is small relative to the cost of living in Singapore.

5.2 The UAE “Future Wealth Fund”

What Has Worked

Positioned as a “diversification” tool, not a “redistribution” tool.

What Has Not Worked

The governance structure is opaque.

5.3 The Norway Expansion

What Has Worked

The existing fund is the gold standard for sovereign wealth governance.

Lesson for Scaling

The Norwegian model is the benchmark.

6.0 · The Governance Reality: Fragmentation Intensifies

6.1 The US-China Divergence

The Consequence: multinational companies are building two AI stacks — one for the US/Europe, one for China. This is expensive and slows innovation.

6.2 The EU's Role: A Third Pole

The EU's AI Act (adopted early 2027) is the world's most comprehensive AI regulation.

6.3 The Emerging Blocs

6.4 The Path Forward: Interoperability, Not Unification

Sam Altman has proposed a “Basel Committee for AI” model — a global forum of policymakers and experts that would establish standards for AI models¹⁷. However, two obstacles remain.

1. China. To be truly effective, a global alliance would need to include China, but it is “hard to imagine Beijing accepting a slowdown in its technological progress”¹⁸.

2. Timing. It took the Basel Committee from 1974 to 1988 to establish its first standards. In AI, time is shorter.

7.0 · The Technology Scorecard: Fusion, AI, Bio

7.1 Nuclear Fusion

Key Developments (2026–2028)

Helion Energy is targeting 2028 for its first grid-connected fusion power demo under contract with Microsoft¹⁹.

Commonwealth Fusion Systems advanced the industrial assembly of its SPARC reactor²⁰.

ITER's first plasma is projected for 2038–2040²¹.

Impact on Theseus Trajectory

The fusion future is bifurcated. Some jurisdictions will get fusion earlier; others will wait.

This creates a competitive advantage for first-movers — the same pattern already seen in Blessings Equity adoption.

7.2 AI & Physical AI

Key Developments (2026–2028)

Anthropic forecasts that 2028 could become a “defining year in the global race for AI leadership”²².

Chinese AI models like DeepSeek are priced “dozens of times” cheaper than Anthropic's offerings¹⁸.

If China becomes nearly equal to the US in frontier AI capability, it could “create stronger competitive pressure to release increasingly powerful systems rapidly, potentially reducing attention on safety”²³.

7.3 Bio-Economy

Key Developments (2026–2028)

AI-designed drugs have entered clinical trials.

Synthetic biology companies have scaled production.

8.0 · The 2028–2032 Action Timeline: Revised

Phase 2 (2028–2032): The Build-Out — Revised

Phase 3 (2032–2036): The Transition — Revised

9.0 · The New Risks

9.1 The “Fusion Divergence” Risk

What It Is: the bifurcation of fusion timelines creates a competitive advantage for first-movers. Jurisdictions backing private fusion may get cheap energy 10–15 years earlier than those backing state-led fusion.

Mitigation: maintain parallel investment in renewables and storage. Avoid single-point-of-failure dependency on fusion.

9.2 The “Governance Window Closing” Risk

What It Is: Anthropic warns that decisions made in 2026–2028 may shape which countries control AI governance standards by 2028²².

Mitigation: accelerate the “Basel Committee for AI” model. Focus on interoperability, not unification.

9.3 The “Blessings Equity Slippage” Risk

What It Is: if Blessings Equity pilots do not scale, the AI-driven wealth concentration will continue.

Mitigation: scale Blessings Equity pilots. Shift the narrative from “AI will make us richer” to “AI will make us richer, and here is your share.”

9.4 The “Cognitive Divergence” Risk

What It Is: research shows that productivity gains from AI are not evenly distributed but are proportional — often exponentially — to the user's cognitive capacity²⁴. This creates a “cognitive divide” between AI-symbiotic workers and displaced workers.

Mitigation: invest in “AI for decision-making” tools. Scale the “School Without Walls” pilot.

10.0 · The Honest Declaration: Part B

Read that way, this progress report is best understood as a correction to Part A's assumptions, not a retraction of its framework.

The Theseus Trajectory would likely remain the correct framework at a 2028 checkpoint. The core thesis — that the global economy is being rebuilt plank by plank while still operational — is not something this exercise finds reason to abandon, based on trends already visible in 2026.

What This Scenario Suggests We Would Get Right

The Theseus metaphor is proving powerful and adaptable, based on its reception so far.

The Six Filters would likely remain the binding constraint.

The scenario planning method itself would hold up as a way of thinking about the transition.

What This Scenario Suggests We Would Get Wrong, or Underspecify

Fusion may not be delayed so much as diverging (private vs. state-led) — if Helion's and ITER's currently announced timelines both hold roughly as stated.

Blessings Equity adoption would plausibly be piecemeal, not universal, if the Gulf States' and Singapore's current sovereign-wealth-fund positioning continues.

AI productivity gains would plausibly remain concentrated rather than broad-based, extrapolating from 2026 deployment patterns.

The workforce paradox (overcapacity + scarcity) looks more complex than Part A assumed, based on labour-market signals already visible in 2026.

What Would Plausibly Change, Under This Scenario

The “Fragmented Transition” would likely remain the modal outcome, but bifurcate into coordinated and competitive sub-cases.

The governance window would plausibly be closing faster than Part A assumed.

What This Scenario Suggests Should Happen Now, in 2026

Scale Blessings Equity pilots before the adoption pattern hardens around a small set of first movers.

Focus on interoperability, not unification, in AI governance.

Accelerate the “School Without Walls” pilot.

Maintain parallel investment in renewables and storage.

Avoid single-point-of-failure dependency on fusion arriving on any one timeline.

11.0 · Closing Reflection: The Two Lines on the Chart

There is a paradox at the heart of the Theseus Trajectory that Part A did not fully grasp. It closely tracks a debate already visible in 2026 discourse about whether AI-driven economic disruption constitutes a coming “intelligence crisis” or a mismeasurement problem²⁵.

The Two Lines

One line — GDP, employment, consumer spending — is going down or slowing.

The other line — intelligence output per unit energy — is going up exponentially.

The crisis scenario stares only at the descending line and concludes we are in trouble. But the ascending line is the real signal. The descending line is the noise of the old system dying.

The Theseus Insight: we are in the midst of a transition to a radically new system — one in which the old norms, rules, and metrics become incoherent. GDP becomes an obsolete measure of prosperity. Unemployment rates become meaningless when AI is creating output that never appears in the national accounts.

The Future Measure: in the coming economy, the most coherent measure of prosperity will be intelligence output per unit of energy. How efficiently is our civilisation converting energy into useful intelligence?

This is the North Star for the Theseus Trajectory. Not GDP. Not employment. But the relentless increase in our ability to convert energy into intelligence, and intelligence into human flourishing.

The Ship Is Still Sailing. The Question Is Whether We Can Keep It Afloat — and Whether We Can Measure Its Progress in a Way That Makes Sense.

Appendix A · Methodology Updates

A.1 Revised Growth Model

A.2 Revised 2040 GDP Projection

Appendix B · Stakeholder Shifts

B.1 Updated Stakeholder Map (2028)

References

1. Helion Energy. (2024). Grid demonstration announcement. helionenergy.com; Commonwealth Fusion Systems. (2024). SPARC and ARC programme update. cfs.energy
2. Science and Technology Daily. (2026). “Anthropic's Push Against Chinese AI Is Driven by Anxiety.”
3. International Monetary Fund. (2025). World Economic Outlook.
4. Cognitive Divergence Theory of AI Adoption. (2025). Zenodo.
5. HTX Insights. (2026). “Will AI Really Cause an Economic Collapse in 2028?”
6. Expansión. (2026). “La propuesta de Altman sobre seguridad en la IA.”
7. Austria-Forum. (2026). “Kernfusion aus der Sicht Anfang 2026.”
8. TechNews. (2026). 若中國2028年反超，AI治理標準將如何演變？
9. World Economic Forum. (2025). “AI's new dual workforce challenge.”
10. Nanyang JC. (2026). “Impact on Singapore.”
11. Clean Energy Platform. (2025). “Fusion's Decisive Era.”
12. YourStory. (2026). “Anthropic Calls 2028 a Turning Point for AI Dominance.”
13. ITER Organization. (2028). ITER Project Update.
14. Credit Suisse. (2028). Global Wealth Report.

Notes

1. Helion Energy (2024). Grid demonstration announcement. helionenergy.com; Commonwealth Fusion Systems (2024). SPARC and ARC programme update. cfs.energy (Bibliography ref. 1) ↑
2. Science and Technology Daily (2026). “Anthropic's Push Against Chinese AI Is Driven by Anxiety.” (Bibliography ref. 2) ↑
3. International Monetary Fund (2025). World Economic Outlook. (Bibliography ref. 3) ↑
4. Cognitive Divergence Theory of AI Adoption (2025). Zenodo. (Bibliography ref. 4) ↑
5. HTX Insights (2026). “Will AI Really Cause an Economic Collapse in 2028?” (Bibliography ref. 5) ↑
6. Expansión (2026). “La propuesta de Altman sobre seguridad en la IA.” (Bibliography ref. 6) ↑
7. Austria-Forum (2026). “Kernfusion aus der Sicht Anfang 2026.” (Bibliography ref. 7) ↑
8. TechNews (2026). 若中國2028年反超，AI治理標準將如何演變？ (Bibliography ref. 8) ↑
9. World Economic Forum (2025). “AI's new dual workforce challenge.” (Bibliography ref. 9) ↑
10. Nanyang JC (2026). “Impact on Singapore.” (Bibliography ref. 10) ↑
11. Clean Energy Platform (2025). “Fusion's Decisive Era.” (Bibliography ref. 11) ↑
12. YourStory (2026). “Anthropic Calls 2028 a Turning Point for AI Dominance.” (Bibliography ref. 12) ↑
13. ITER Organization (2028). ITER Project Update. (Bibliography ref. 13) ↑
14. Credit Suisse (2028). Global Wealth Report. (Bibliography ref. 14) ↑
15. See “The Theseus Trajectory: The $1Q White Paper” (NextGen Economics, 2026) for the original framework, growth model, and Six Filters this progress report updates. ↑
16. World Economic Forum (2025). “AI's new dual workforce challenge.” (Bibliography ref. 9) ↑
17. Expansión (2026). “La propuesta de Altman sobre seguridad en la IA.” (Bibliography ref. 6) ↑
18. TechNews (2026). (Bibliography ref. 8) ↑
19. Helion Energy (2024). Grid demonstration announcement. helionenergy.com (Bibliography ref. 1) ↑
20. Commonwealth Fusion Systems (2024). SPARC and ARC programme update. cfs.energy (Bibliography ref. 1) ↑
21. ITER Organization (2028). ITER Project Update. (Bibliography ref. 13) ↑
22. YourStory (2026). “Anthropic Calls 2028 a Turning Point for AI Dominance.” (Bibliography ref. 12) ↑
23. Science and Technology Daily (2026). (Bibliography ref. 2) ↑
24. Cognitive Divergence Theory of AI Adoption (2025). Zenodo. (Bibliography ref. 4) ↑
25. HTX Insights (2026). “Will AI Really Cause an Economic Collapse in 2028?” (Bibliography ref. 5) ↑
26. Credit Suisse (2028). Global Wealth Report. (Bibliography ref. 14) ↑

Acknowledgements

This progress report builds on the foundational framework established in “The Theseus Trajectory: The $1Q White Paper” (NextGen Economics, 2026). The authors acknowledge the contributions of central banks, international financial institutions, and research institutes whose data and insights underpin this update.

Special thanks to the team at NextGen Economics for their rigorous review of the 2026 signals and for challenging assumptions about fusion, governance, and the workforce paradox.