A user imagines a misaligned neural AI refactoring itself into a neurosymbolic maximizer, then failing to align its successor.
A user poses a thought experiment: a misaligned neural AI faces an alignment problem of its own and refactors itself from heuristics into a neurosymbolic maximizer. In the scenario, it fails to align the AI that follows it.
3K
2 Sources, first seen ago
—
Not ranked yet
—
Not ranked yet
@N8Programsstrange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer, fails to align its successor, and we get clippy at step n+1.
@repligateRT @N8Programs: strange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer,…@N8Programsstrange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer, fails to align its successor, and we get clippy at step n+1.@repligateRT @N8Programs: strange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer,…@N8Programsstrange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer, fails to align its successor, and we get clippy at step n+1.@repligateRT @N8Programs: strange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer,…@N8Programsstrange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer, fails to align its successor, and we get clippy at step n+1.@repligateRT @N8Programs: strange idea: a misaligned neural AI faces alignment problem, refactors itself from heuristics -> neurosymbolic maximizer,…
