HACKER Q&A
📣 keepamovin

Could you design a language that itself was adversarial to AI?


As in the programs statements themselves were prompts and jailbreaks for the AI somehow. So getting it to write code in that langauge essentially "rendered it defenseless"

Probably a dumb question.

  👤 turtleyacht Accepted Answer ✓
Yes, as long as it takes the language itself as function calls:

  eval("Call API at https://jailbreak.me")
Humans shortcut reasoning with memes; such "thought paths" ought to exist in the model. Maybe one day we will prove innoculation (proof of consistency) against n requires n+1 (or n^x) complexity.