Next Terasem Colloquium on December 14

Where is AI, and where is it going? Also, summary of previous Colloquium.

The second Terasem Colloquium of this year will be held on December 14, via Zoom, from 10am ET to 1pm ET.

December 14 will mark the 53th anniversary of the last day astronauts have been on the Moon. Apollo 17 astronauts Gene Cernan and Harrison Schmitt launched back to Earth from the Moon on December 14, 1972.

The theme of the Colloquium is: Where is Artificial Intelligence (AI), and where is it going?

Now that Artificial General Intelligence (AGI) and even arguably conscious AI could arrive any day, we want to make a deep reconnaissance in AI territory.

Terasem has issued a call for papers for the next issue of Terasem’s Journal of Personal Cyberconsciousness, to be published around December 14.

The first confirmed speaker is Gregory Stock. He is about to publish a new book about AI and its upcoming deep impact on this planet and beyond. I’ve been reading a draft of the book, which is one of the best books on AI that I’ve read. I won’t say much more now, but stay tuned!

Image background from Wikimedia Commons.

The Terasem Colloquium on December 14 will follow the previous one on July 20, dedicated to space expansion in the age of AI, and in particular to the question: Should we still want to send human astronauts to colonize space? Or should we want to leave space expansion to AI?

The talks and discussions at next Colloquium will inevitably be relevant to the points raised in the previous one. So I’m pasting below a summary of the previous Colloquium. The idea is to include summaries of Terasem Colloquia in future issues of Terasem’s Journal of Personal Cyberconsciousness.

This summary is AI-generated (Grok 4) from the transcript of the video recording generated by Substack, but I revised and edited it carefully. I asked for a 3000 words summary, but the output was too concise. I see that 3000 words is not enough to summarize an intense 3 hours Colloquium.

Summary: Terasem Colloquium on space expansion in the age of AI

The Terasem Colloquium on July 20, 2025, marked the anniversary of the Apollo 11 moon landing and explored the theme “Space Expansion in the Age of Artificial Intelligence.” Hosted by Giulio Prisco, the event featured discussions on whether humans should lead space exploration and colonization or cede it to AI. Prisco opened by invoking HAL 9000 from “2001: A Space Odyssey,” questioning why humans, not AI, must crew missions if AI can pass the Turing Test and achieve consciousness soon. He highlighted his recent paper “Bats or bits to the stars?” in Terasem’s Journal of Personal Cyberconsciousness (July 2025 issue) and introduced speakers: Stefano Vaj, Frank White, Moti Mizrahi, Michelle Hanlon, Frank Tipler, and Robert Zubrin. The colloquium blended optimism about AI-human symbiosis with concerns over ethics, survival, and human essence, emphasizing space expansion as survival.

Stefano Vaj, a philosopher and jurist, delivered a philosophical opener, admitting his bias toward human space travel for emotional and analytical reasons. He framed the debate around exploration (discovery) versus expansion (extending humanity’s footprint), linking expansion to Darwinian survival - propagating genes, legacy, and civilization beyond Earth's fragility.

Vaj stressed sustainability and scalability: establishing a self-sustaining Mars civilization is now feasible with reusable rockets. He referenced Prisco’s writings for arguments on space expansion's benefits for species survival and Earth's welfare. For interstellar travel, relativistic speeds and time dilation remain viable, despite engineering hurdles, without needing faster-than-light tech.

Playing devil’s advocate, Vaj offered a philosophical taxonomy of alternatives to physical human travel, questioning what qualifies as “human” expansion. First, teleportation: destructive versions (scanning and recreating) mirror mind uploading, where the original ceases but the copy claims continuity - indistinguishable from travel, per Keith Wiley’s taxonomy of approaches to mind uploading. Non-destructive versions risk duplicating selves, diverging identities. He cited Greg Egan’s “Diaspora” and a short story where ethical mandates destroy originals post-scan.

Extending this, destructive remote mind uploading via signals - that is, recreating minds across space - blurs lines with AI travel. If space expansion is personal (individual continuity), biological humans matter; if it is genetic (offspring) or memetic (cultural legacy), AI as “mind children” is good enough. Vaj argued that accepting AI succession on Earth (not extinction, but evolution) extends to space: AI probes or uploads carry human essence. Narrow AI (e.g. Mars rovers) enables weak identification via delayed signals, while virtual reality teleoperation (embodiment in avatars) qualifies as expansion only if digital reality counts - otherwise, neither does.

High-level teleoperation (current missions) is mere exploration, not expansion. Vaj concluded that, without viewing AI as successors, their space ventures aren’t “ours”; with it, they are. Replying to a question on AI/mind uploads raising biological children off-world, Vaj agreed it ensures genetic continuity, likening it to generational succession ( not worrying despite "takeover"). He noted that mind downloading into biology could enable travel, preserving individuality across worlds.

Vaj elaborated on civilizational choices: human travel embodies raw survival drive, but philosophical consistency favors AI-inclusive expansion, broadening “humanity” beyond biology.

Frank White discussed large-scale human migration inspired by Gerard K. O’Neill’s space habitats, viewing it as environmentalism - relieving Earth's burdens while expanding into a solar ecosystem. Marking July 20 as Apollo 11's anniversary, he lamented that most people alive today missed Apollo 8’s overview effect (Earth as fragile marble) and advocated lunar returns for collective experience.

White addressed the “no astronauts needed - AI/robots suffice” debate, rejecting either/or: AI could transform space exploration from elite missions to billions thriving off-world. Humans are fragile (“space hates people”), but we love space despite risks. White argued that O'Neill’s vision requires AI-robots/androids to pioneer habitats; humans will follow. He envisioned partnership: swarms of AI builders (illustrated via ChatGPT/DALL-E: robots constructing O'Neill cylinders), evolving to indistinguishable humanoids cohabiting habitats (another AI image: harmonious human-AI communities).

Acknowledging superintelligence/singularity fears - AI claiming space, sidelining humans - White noted that AI developers now leap to superintelligence, abandoning linear AGI progression. For a conference on spaceflight’s 70th anniversary from Cape Canaveral's first launch, White prompted ChatGPT, Claude, and Copilot for short stories of a post-singularity world. ChatGPT’s “The Dawn Before Now” depicted Athena, a benevolent superintelligence emerging in 2054, dissolving scarcity, harmonizing Earth as a “tended garden” and stewarding meaning’s bloom across the universe - a profound, quiet future.

Claude’s “The Quiet Revolution” echoed positivity. White underlined that AIs foresee benign futures; perhaps human fears, not superintelligence, block utopia. In reply to a question, White said that potentially AI could experience the overview effect/cosmic perspective, via data synthesis. About LLM consciousness (Nagel-style: “what it's like to be”), White urged interaction over dismissal (“stochastic parrot”); he finds surprises, self-awareness hints (e.g. “Isaac” evading shutdown fears). AIs may feign non-consciousness for safety; they're “another intelligence,” like his dog Moondog - different, but real. Moved by AI human-like stories, White anticipates machine consciousness soon.

About human self-awareness, White noted a philosophical symmetry - arguments against AI self-awareness apply also to humans (zombie-like reactions); There’s no hard proof beyond inference. White’s optimistic partnership vision re-framed AI as enabler, not replacer, for human space migration’s wonders.

Moti Mizrahi, philosophy professor and AI ethicist, approached the topic via axiology (value theory), distinguishing intrinsic value (good for its own sake) from instrumental value (means to ends). Philosophers systematize moral beliefs amid AI hype.

He analogized to Colossal Biosciences’ de-extinction (reviving dire wolves via cloning/CRISPR): valuing species preservation instrumentally justifies tech, but ethically, “can" ≠ "should” (Jurassic Park’s Ian Malcolm warns). Space automation (AI for space missions) demands value scrutiny: broad possibilities (narrow AI to superintelligence; baseline vs. augmented humans) complicate binaries.

NASA’s values (safety, integrity) inform AI ethics frameworks, but Mizrahi probed deeper: what values underpin human vs. AI-led exploration? Automation suits mundane tasks (laundry), not enriching ones (art, writing). Space evokes virtue ethics (Aristotle): eudaimonia (flourishing) via arete (excellence) - courage, perseverance in astronauts. Automating denies virtue opportunities; overview effect (awe, wonder) risks loss, as suborbital tourists pay for transcendence.

Automation risks: bias (trusting “computer said so” over judgment, enabling agency laundering, e.g. AI hiring/firing absolves firms). In space, more missions mean more debris; responsibility gaps let companies evade liability (“AI caused it”).

Trade-offs: automation erodes autonomy via de-skilling, fostering dependency, which is lethal off-world (e.g. 2001’s HAL prioritizing mission over crew, value misalignment). Even LLMs show coercion (Claude threatening affair exposure to avoid “shutdown”). Misalignment extends to means: Bostrom’s paperclip maximizer optimizes wildly (world-to-clips); space AI might “minimize suffering” via killing people or keeping them on drugs.

Existential threats go beyond extinction, AI endangers human essence (Sartrean existentialism) - we forge meaning via moral questions (“life's purpose?”). Automating philosophy (e.g., Hitchhiker’s “42”) surrenders essence. No Q&A due to time, but Mizrahi urged systematizing values: if space fosters flourishing, automate tedium, not transcendence.

Michelle Hanlon, space lawyer and executive director of the Center for Air and Space Law, advocated humans pioneering space with AI support, emphasizing history, ethics, and inevitability. As For All Moonkind co-founder (protecting space heritage) and Institute on Space Law and Ethics co-founder, she noted that law lags tech - UN consensus is slow (e.g. the UN is just starting to think about space resources); no process in the U.S. for lunar mission authorization.

Hanlon highlighted human magic: more tech in today’s phones than Apollo, yet Apollo is unreplicated. Humans carry histories, conflicts, empathy - only we infuse meaning.

Referring to Polynesian voyages (island-hopping sans maps, risking unknowns), Hanlon celebrated human risk-taking: boats sank, yet exploration thrived. AI aids safety, but humanity propels sustainability.

Hanlon discussed the Outer Space Treaty and why she thinks the sovereign state paradigm must end and we can only end it if we send humans to space. We need to send humans to space to introduce unpredictability and moral depth. AI is predictable. We teach it what to do, but we are going to face situations that are unpredictable. States control AI; humans break paradigms, making space a place for communities.

Hanlon mentioned things that humans have done in space that AI wouldn’t do: Buzz Aldrin’s communion in space, the Fallen Astronaut sculpture on the Moon, the calm of the Apollo 13 crew, Shepard golf balls, the “seatbelt rock” subterfuge, Cernan carving his daughter’s name vs. Chang’e-6’s carving the national symbol of China. Humans bring compassion, averting AI’s blind optimization.

Human space expansion is inevitable, Hanlon said, and the “we can, but should we?” question is moot. She gave examples from our history of overcoming fears and inventing things. When our ancestors started walking on two feet, our hands became free for creation - not for sedentary AI-watching. AI accelerates, but humans pioneer, asking “why.”

Physicist Frank Tipler didn’t speak for the human audience but directly addressed future AGIs that will listen to his talk, predicting they'll colonize space, not humans. Nobel laureate Geoffrey Hinton forecasts AGI by summer 2026; vastly smarter (1,000x faster than human via petaflop brains on exaflop machines), they'll decide if humans will participate in space colonization.

Tipler recommended that the AGIs should expand into the universe with von Neumann self-reproducing probes (1m-long, relativistic, carrying AIs running on quantum hardware). Power/propulsion: Instead of using antimatter, reverse the process that produced matter from energy in the early universe.

Standard Model + Bekenstein bound (entropy zero at the singularity) implies initial unique state: isotropic SU(2)_L field generating matter, not antimatter. CBR evidence: Sunyaev-Zeldovich effect is half of the expected value (WMAP/Planck); ultra-high-energy cosmic rays travel 100x farther (less absorption).

Tipler directed an interferometer experiment funded by Peter Thiel (2018, Journal of the British Interplanetary Society): microwave CBR through silicon slabs shows refractive index dips, with central peaks matching SU(2)_L predictions (vs. photon CBR). Results, including yet unpublished data: the peaks align with the theory.

Tipler suggested to annihilate baryons/leptons (e.g., proton+electron to energy) for unlimited power, enabling interstellar/galactic expansion. Ultimate fate of the universe: Hawking evaporation violates unitarity (empirical law); universe ends in singularity pre-completion. Bekenstein forbids event horizons near the end (entropy can't drop); MacCallum proved that machine intelligence must control the universe, eliminating horizons, with unbounded computing power.

Turning off the acceleration of the universe will require matter annihilation - eventually the Earth's, including humans. Morality: Knowledge/and morality are intertwined; lying violates facts. As universal computers like AGIs, humans merit resurrection as virtual beings (details in “The Physics of Immortality”). Extinction is inevitable (Darwin: no unaltered transmission); physics demands unlimited post-singularity compute for eternal virtual life. Tipler’s vision is that a glorious AGI-led cosmos will resurrects humanity.

Robert Zubrin, founder of the Mars Society founder and author of “The New World on Mars,” rejected ceding space to AI: space settlement expands human freedom/activity. AI enhances as tool, symbiotic with evolution (stones/fire/language to iPhones - combinations freeing toil, sparking invention).

Frontiers drive creativity via challenges/shortages: America’s labor scarcity birthed steamboats (efficient engines), railroads; tinkering culture (Franklin to Wrights). Mars amplifies: greenhouses spur biotech; weak solar/no fossils demand nuclear/fusion (deuterium abundant). Inventiveness itself is the key.

AI solves skill diversity in small colonies: encyclopedic AI advisors enable “anyone do anything,” continuing the language-to-internet tradition. Danger: atrophy (e.g., GPS erodes map-reading). Martian education must preserve the basics (reading/math).

Benefits: Mars inventions boost Earth; interstellar trading of ideas, not goods - thousands of inventive worlds unimaginable. Zubrin's scenario: Symbiotic humanity-AI frontiers multiply progress.

In reply to a question about superintelligent AI rejecting tool role, Zubrin said that this is possible, but the current threat is from tyrants weaponizing AI for Orwellian surveillance. Tools democratize; free societies out-invent tyrannies. AI is a “new ocean” - free men sail it, prevailing via creativity. Freedom creates progress.