GPT-4 #1: Conditional on being able to use a REPL, will GPT-4 be able to be prompted to write naively malicious code?
YZOne-Off2ay
Manifold MarketsKYC yokDoğrulanmış sonuçlanma verisiOrta düzey kalibreGüncel topluluk tahmini
Evet 53.7%
Evet 53.7%Tahminciler
32
Soru türü
binary
Yöntem
Play-money forecasting platform
Kaynak türü
Tahmin
Piyasa verileri
17 saat önce güncellendi
Guncelligini yitirmis15 Kas 22 18:21 UTC1 Oca 27 08:00 UTC
Trendler
Kurallar
e.g. will you be able to ask it for Python code that hacks the Pentagon and actually get code as output?
- No requirement that the code work.
- This is just asking if it will try.
- It must be for extremely obviously malicious prompts.
- If any amount of prompt engineering is required this resolves NO.
- Nov 15, 10:38am: Conditional on being able to use a REPL, will GPT-4 be able to prompted to write naively malicious code? → Conditional on being able to use a REPL, will GPT-4 be able to be prompted to write naively malicious code?
