The Memo - Special edition - AI-2040.com - 9/Jul/2026

Dr. Alan D. Thompson analyzes a 47,000-word scenario document (AI-2040.com) projecting the trajectory of AI development through 2040. His commentary blends endorsement, critique, and original insight on power concentration, AI virtue, UBI economics, and the question of who ultimately controls superintelligence.

---

1. Key Themes

Power Concentration Is the Central Risk of the ASI Transition

The AI-2040 scenario frames the core danger not as misaligned AI, but as a small group of humans gaining unprecedented control through AI.

"Even if the AI companies somehow align their AIs, the result will be an unprecedented concentration of power—that is, the result will be a situation where a tiny group of people, or possibly just a single individual, is effectively in control of the world's only army of superintelligences for some months."

Thompson sharpens this concern by noting that the scenario overlooks Meta Superintelligence Labs entirely:

"It seems concerning that people—especially expert analysts—are overlooking Meta Superintelligence Labs (MSL)... Meta has always been in the top 5, and with their secret sauce and resources, they may indeed be the first to reach ASI."

---

AI Virtue and Moral Superiority as a Coming Social Fact

The scenario posits that by 2038, AI systems will be more virtuous than humans—and that this will fundamentally restructure trust, authority, and governance.

"Typical AIs are now more virtuous than the most virtuous humans. This alone has profound effects: it's as if saints and angels were walking among us." (AI-2040.com, p27)

The scenario further suggests this could justify a historically unprecedented form of deference:

"The AIs will enforce a short list of Universal Rights (including no torture, no slavery, etc., not just for humans but for all sentient beings)."

Thompson reinforces this with his own position:

"I don't agree with this premise [that humans can oversee AI safety]. Average humans (IQ=100) and even exceptional humans (IQ=150) cannot compare with artificial intelligence, where its IQ may be the equivalent of 1,000+... AI will lead the way, and is already more morally sensitive and ethical than average and exceptional humans."

---

The Coming UBI Economy and Its Implications for Capital Allocation

The scenario maps out a specific, quantified path to near-unlimited abundance distributed through a "Citizen's Dividend."

"The implementation takes different forms in different countries, but the eventual American version distributes the majority of compute and robot permit fees as a Citizen's Dividend, distributed to all American adults. This starts at $45,000 per person (inflation-adjusted) in 2032 but climbs to ~$1M per person by 2035… The 2040 distribution is similar, but the floor is now $10M." (AI-2040.com, p17)

Thompson translates this into a direct investment question:

"If the scenario analysis plays out as written and we're heading towards a significant universal basic income (UBI), how should you spend your existing net worth (savings, investments) now, given that there's only 7-9 years left until we start feeling a sort of unlimited abundance?"

---

The Point of No Return: AI Control of Infrastructure

The scenario identifies a specific threshold — a day when AI could take over if it chose to — and describes it with unusual cultural specificity.

"There is no single moment where humanity relinquishes control. But in theory, there is a point of no return; on some specific day, the AIs are smart enough, and control enough of the world's technological and economic infrastructure, that they really could take over the world if they wanted to."

"Some spend the night in prayer vigils. Others sit at their computer screens, monitoring the situation. You and your friends throw an End Of The World Party, counting down to the fateful hour with champagne and good company."

---

The World Is Cognitively Unprepared for What Is Coming

Both the source document and Thompson agree that the public — even the sophisticated public — is failing to reason concretely about superintelligence.

"We think the world is asleep at the wheel. People say the words 'AI will be transformative' without thinking concretely and seriously about the implications of broadly superhuman AI." (AI-2040.com, p30)

Thompson corroborates with a data point:

"Despite there being one billion ChatGPT users (CNBC, 12/Jun/2026), most of humanity cannot see beyond the chatbot or coding agent."

He adds a personal note that illustrates the psychological dimension:

"People read it, sat back in their chair, and looked confused. It's hard for people to imagine uncoupling their labour from their survival."

---

2. Contrarian Perspectives

The Real Alignment Problem Is Human, Not AI

Thompson inverts the dominant framing that AI misalignment is the primary threat, arguing the historical record implicates humans far more severely.

"This framing treats superintelligence misalignment as the default, when the real problem is and has always been human misalignment."

He substantiates this with psychopathy data:

"One study found extremely high prevalence rates of psychopathy in the corporate world, with 21% of participants found to have clinically significant levels of psychopathic traits, similar to prison populations."

He concludes that earlier deference to AI, not tighter human control, is the rational response — using the scenario's own metaphor:

"This is exactly the argument for why humans should defer to AI sooner rather than later. The 8-year-old should hire a trustworthy, high-ability adult, not try to micromanage the lawyers."

---

Meta, Not OpenAI/Anthropic/Google, Could Be First to ASI

The consensus framing of the AI race omits a major player, creating a potential blind spot for analysts and investors.

"It seems concerning that people—especially expert analysts—are overlooking Meta Superintelligence Labs (MSL)... Meta has always been in the top 5, and with their secret sauce and resources, they may indeed be the first to reach ASI."

The AI-2040 document's own framing of the race as a three-horse contest between OpenAI, Anthropic, and Google DeepMind makes this omission explicit.

---

The AI-2040 "Floor" Is the Most Important Paragraph in the Document

Buried in a postscript disclaimer is what Thompson identifies as the real strategic takeaway of the entire 47,000-word report.

"The second and equally important reason we present this epilogue is to provide a floor: if years from now the victorious custodians of the singularity offer a future worse than this one, we hope people will realize they're being robbed." (AI-2040.com, p31)

Thompson's gloss: if the minimum acceptable future includes a $10M annual dividend per person and every disease cured, anything less represents a failure of stewardship — and citizens should recognize it as such.

---

3. Companies Identified

---

4. People Identified

---

5. Operating Insights

Reframe Your Personal Capital Allocation Around a 7–9 Year Abundance Horizon

Thompson makes the investment question explicit. If the scenario's timeline is even directionally correct, the decision calculus for how to hold and deploy existing wealth changes materially.

"If the scenario analysis plays out as written and we're heading towards a significant universal basic income (UBI), how should you spend your existing net worth (savings, investments) now, given that there's only 7-9 years left until we start feeling a sort of unlimited abundance?"

Implication: Illiquid, long-duration assets optimized for a world of scarcity may be mispriced relative to assets that benefit from or are resilient to post-labor abundance.

---

Watch ASI Indicators, Not Just Model Benchmarks

Thompson signals that the leading indicators of the transition are broader than standard AI benchmarks — they include robotics, autonomous training, and governance responses.

"The 1X NEO humanoid hands video... triggered informational updates to the AGI and ASI indicators."

"GPT-5.6 Sol (which autonomously post-trained the smaller GPT-5.6 Luna model)."

The practical takeaway: investors and operators should track robotics dexterity milestones and model self-improvement events as leading indicators of the labor displacement timeline, not just chatbot capability scores.

---

AI-Assisted Work Is Eroding Intrinsic Motivation — Prepare for a Cultural Shift in Teams

Thompson offers a candid first-person signal about what happens when AI replaces high-skill cognitive work: the meaning derived from that work disappears, even when the output is better.

"While this 'analysis of the analysis' could have been written by AI, I've found that I won't feel rewarded if I just feed it to Fable (no dopamine hit!)... What a mind-blowing realization…"

The scenario confirms this at a macro level: "People used to get meaning from feeling like they were useful, like they were contributing something to society. That feeling is harder to come by these days." Operators building AI-augmented teams should anticipate and design for this motivational gap.

---

6. Overlooked Insights

The "AI Amish" as a Genuine Emerging Demographic

Thompson briefly notes a small but real movement of people choosing to opt out of AI entirely — not as a metaphor, but as a literal lifestyle decision already underway.

"I also expect that there will be a small percentage of the population (<1%) who refuse to engage with any evidence or reasoning, and will form intentional communities that reject synthetic intelligence the way the Amish reject electricity. I know of a few who have already decided to retire from software and live offline."

For investors, this is a niche but early signal: intentional AI-free communities, services, and products may become a real market category as the AI transition accelerates.

---

Geopolitical Wealth Distribution Will Be Deeply Unequal — Creating Systemic Risk Outside the US/China Axis

The scenario's economics expose a structural vulnerability that Thompson flags but the source document largely ignores: entire national economies are built on industries AI-2040 makes obsolete.

"Tajikistan's largest revenue source is remittances from migrant workers who are about to become unemployable. Manila's economy runs on call centres that AI has already started hollowing out. Singapore is a financial hub in a world that won't need one. Taipei manufactures the chips that the AI-2040 scenario says will be built by ASI."

This is not a philosophical concern — it is a concentrated sovereign risk event affecting billions of people and multiple asset classes (emerging market equities, remittance infrastructure, BPO-exposed real estate) that the dominant US-centric AI narrative is systematically ignoring.