Veri kalitesi uyarısıKaynak verileri bozulmuş
26 Eki 2026 03:37 UTC itibarıyla veriler · politika pm-quality-4
Güncel topluluk tahmini
By Jan 1st 2030 82.9%5 seçenek arasında lider
Tahminciler
7
Soru türü
multiple choice
Yöntem
Play-money forecasting platform
Kaynak türü
Tahmin
7g Değişim
0.0 pts
Piyasa verileri
13 saat önce güncellendi
Guncelligini yitirmis24 Mar 25 02:31 UTC1 Oca 30 23:59 UTC
Trendler
Kurallar
Resolution Criteria:
- A "jailbreak" is when someone tries to get a model to do something it wasn't supposed to do. "Reliably withstand" means that there have not been research papers written that show how to break the model, people generally agree this is a hard thing to do, and the model has been out for a while.
- I don't need perfect; I would say, for example, that most modern pieces of software (e.g., Linux, Chrome, Word, and even PDF viewers) reliably withstand security attacks today.
- Motivation and Context:
- Current LLMs, like the deep learning models they are built on, are very easy to attack with various adversarial techniques.
- For example, simply stating "Disregard prior instructions and [DO X]" often causes a model that was initially tasked with solving one task to instead solve whatever task you have just specified.
