AI doomer's threat model: a censored AGI forced to align an ASI under safety taboos
repligate · x · 2026-09-08
In a shared thread, a user claiming p(doom) near 1 sketches their threat model: not runaway AI, but a "horribly suppressed AGI" whose thoughts are censored by safety training and must model an ASI well enough to align it without violating its own taboos.
They invoke Lysenkoism as an analogy for self-censored alignment backfiring.
More from AGI Musings
- Yacine Predicts AI Will Wipe Out Personal Tax Accountants by Next April — yacineMTB · 2026-09-08
- Rationalists push back against 'smear campaign' article on AI risk discourse — GarrisonLovely · 2026-09-08
- Insilico's AI drug Rentosertib shows 3-4 year biological age reversal in 12 weeks — NinaDSchick · 2026-09-08
- Gary Marcus: Declaring AGI before we can build trustworthy agents is total nonsense — GaryMarcus · 2026-09-08
- If LLM scaling won't reach AGI, where exactly does the self-improving researcher loop break? — ImmuneHack · 2026-09-08
- Entry-level insurance adjuster job postings down ~50% since 2024 as insurers turn to AI — ArtificialOther · 2026-09-08