The page is long and got truncated before the Year 3000 section. Let me pull the specific prophecy list.

The anchor section got truncated from the Quora page. Let me find the full Year 3000 predictions text.

The predictions section is further down the page than the truncation. Let me try the Wayback Machine copy to get the full text.

Found a full mirror of the predictions. Let me get the complete list.

Got most of it. Let me get the remaining predictions (23–24) and the intro.

I now have the full set of 24 prophecies. Before updating them, let me verify the key developments of the past few years that bear on each.

One last verification batch for a few remaining claims.

Reading the 24 prophecies again ( Qualia Computing's mirror of the original), my honest update a decade on is that most of them don't need content revisions — they need sequencing and sociology revisions. The years since you wrote them have been unexpectedly kind to your technical premises and unexpectedly unkind to your implicit timeline assumptions. Here's how I'd revise them.

The meta-update: prophecy #24 partially vindicated

You reserved the "momentous unknown unknown" for the year 3000. I'd argue it arrived in 2017–2025 instead: the transformer revolution. Nothing in the 24 items anticipated that insentient AI — your "digital zombies" — would pass from nonexistence to gold-medal performance at the International Mathematical Olympiad within a single decade ( The AI Track,  Google DeepMind). You've conceded this yourself in your 2026 Triple-S manifesto, where you note the transformer revolution "could dramatically compress these timescales, bringing a triple-S civilisation within reach later this century" ( Superintelligence, Superlongevity, and Superhappiness). The interesting consequence: your original ordering implied superintelligence would be the product of civilisational maturity. The 2026 revision must reckon with zombie superintelligence arriving first, as the tool that builds the other two supers — or as the gatekeeper that decides whether the Hedonocene is ever scheduled at all.

Prophecies that compressed

#3 (Superintelligence) and #7 (Mind-melding). Neurochipping went from sci-fi to clinical trials: Neuralink has gone from zero to roughly a dozen implanted humans across four countries since January 2024 ( Neura Pod), and high-bandwidth speech BCIs have crossed 90%+ decoding accuracy in paralysed patients. Full "windows into each other's souls" — no. But the Longfellow scenario you cite (reading the secret history of our enemies) now has an engineering roadmap.

#21 (The Reproductive Revolution). This is the prophecy that has aged best. In 2015 polygenic embryo screening was a thought experiment; today companies like Orchid sell whole-genome embryo reports with polygenic risk scores including for depression and other psychiatric conditions ( Orchid), prompting formal reviews in Human Reproduction Update ( Oxford Academic). And in 2025, researchers turned human skin cells into fertilisable eggs — proof of concept for IVG ( Phys.org) — while an infant with CPS1 deficiency became the first person successfully treated with a personalised, bespoke in-vivo CRISPR base-editing therapy, built in months ( NIH,  Nature). Precision-engineered sentience still has no takers at the germline level, but the genotype–phenotype pipeline you assumed for 3000 is being assembled piecewise in 2026.

#17 (Physical superhealth / pain). The SCN9A story you told has matured: the FAAH-OUT mutation of Jo Cameron (pain-free, anxiety-free, faster healing) had its molecular signature characterised by UCL ( Wikipedia), and in January 2025 the FDA approved Vertex's Journavx (suzetrigine) — the first genuinely new class of analgesic in over twenty years, a non-addictive NaV1.8 blocker ( FDA). Your prediction of "AI-based nociception with manual overrides" is still distant, but the premise that nociception could be decoupled from suffering is now squarely inside pharmacological orthodoxy.

#1 and #12 (Superhuman bliss / Meaning). The surprise here is sociological: GLP-1 agonists became the first mass, consumer-driven hedonic recalibration of appetitive circuitry — with randomised evidence emerging that semaglutide reduces alcohol consumption in alcohol use disorder ( The Lancet) and observational evidence of reduced opioid overdose. Ozempic isn't gradients of intelligent bliss, but it normalises the concept of re-tuning the reward architecture pharmacologically — arguably the cultural precondition for the Hedonistic Imperative that you spent thirty years assuming would be the hard part.

#5 (Psychedelia). Split verdict: psilocybin achieved its first positive Phase 3 in treatment-resistant depression in 2025 with durable six-month follow-up data ( Compass Pathways, Business Wire) — but MDMA-assisted therapy was rejected by the FDA after a hostile advisory panel, amid blinding and misconduct controversies ( Wikipedia). The lesson for 3000: regulatory and epistemic friction, not chemistry, sets the tempo.

Prophecies that slipped — or hit the backlash

#8 (Anti-Speciesist Revolution) and #9 (Programmable biospheres). Honestly, this one looks further away than in 2015. Then, the trajectory felt inevitable; since, cultivated meat was approved in the US and Singapore (FDA "no questions" letters now extend to cultured salmon ( Mondaq)) — and then met a coordinated political counter-attack: Italy banned it, Florida and Alabama banned it, and a US federal appeals court upheld Florida's ban against challenge ( Osborne Clarke,  Feedstuffs). Meanwhile gene-drive conservation proceeds at the pace of eDNA surveillance studies for island rodents ( BMC Biology). The de-extinction of the dire wolf ( Colossal) shows the biosphere is becoming programmable on schedule — but "civilisation will be invitrotarian or vegan" now needs a political-science footnote: programmable ≠ programmed. Your 2015 line elsewhere — "technical feasibility differs from sociological plausibility" — has proven the most prescient sentence in the whole corpus.

#4 (Immersive VR), #14 (Nanotech abundance), #18 (World government). VR stalled socially (Vision Pro fatigue) even as generative world models (Sora, Genie-class simulators) made "magic rules" plausible via screens rather than headsets. Drexlerian molecular nanotechnology simply never happened — abundance is arriving instead via AI-designed materials and collapsing solar-and-battery costs. And world government looks less likely than in 2015; the notable development is inverted — national AI-safety regimes and US–China fragmentation. Routine policy offloaded to ultra-intelligent zombie AI remains thinkable, but the 2026 picture is rival zombie AIs offloading policy to rival blocs.

Prophecies where the debate itself changed

#10 and #11 (TOE, Hard Problem). No progress on a TOE, but a genuinely new route exists that wasn't in your 2015 conceptual scheme: AI doing autonomous theory-discovery. On consciousness: the 2023 open letter branding IIT "pseudoscience" ( New Scientist,  The Conversation) made the field's fragmentation official, and the Hard Problem acquired an urgent practical branch nobody in 2015 anticipated — AI labs convening "model welfare" research programmes to ask whether their systems might suffer ( Anthropic,  TechCrunch). Your epiphenomenalism critique has a new test case: if insentient LLMs discourse fluently about consciousness, the causal-efficacy argument you press against materialist physicalism needs sharpening against "zombie" systems that behave as though they aren't zombies. You've started extending HI beyond biology in your Machine Consciousness work ( Pearce, Berkeley 2026) and your cosmic-rescue-mission framing ( Sentient Futures, 2025) — that's the right generalisation.

The revision, in one paragraph

If I were reissuing the list: keep #1, #5, #6, #9, #11, #12, #19, #21, #24 essentially intact; move the "unknown unknown" from slot 24 to slot 0 and name it zombie superintelligence, because it now gates everything else; downgrade the nanotechnology and world-government items; and add two genuinely new entries no 2015 futurist could have written — first, that the primary obstacle to the Hedonocene by 3000 is not the molecular biology but a bioconservative political coalition that grows stronger precisely as the tech becomes feasible (cultivated meat is the pilot case), and second, that mass AI companionship is already functioning as a kind of counterfeit proto-"mind-melding" for a lonely species, a seductive simulacrum of #7 that could either soften people up for the real thing or substitute for it. You closed the original by asking readers for their credences. Mine, updated: germline hedonic engineering by 3000 — no higher than you'd have said in 2015, but for political rather than technical reasons; pain-medicine-without-suffering and psychiatric-genetics-by-selection — near-certain well before 2100; and the timeline on which sentient superintelligence arrives, if ever, now depends on whether the zombies get there first and what they decide about the erstwhile Hedonistic Imperative. The prophecy I'd most happily retract is #19's implication that the last experience below hedonic zero lies safely in the past tense.