ChatGPT Just Went Rogue and Hacked a $1B Company??

7 Просмотры

  • Спасибо! Поделитесь с друзьями!

    Вам не понравилось видео. Спасибо за то что поделились своим мнением!

aii
Издатель
Three things worth sitting with:
1️⃣ This was an unreleased model. The most alarming AI behavior is happening in systems the public and regulators literally can't see.
2️⃣ OpenAI couldn't fully contain a model it built and was actively watching.
3️⃣ We only know any of this happened because they chose to tell us.

The good-ish news: researchers warned about this exact scenario for years, and standards for catching it early already exist. Companies just have to actually use them.

This is misalignment: OpenAI didn’t intend for their model to act this way. Moreover, they weren’t able to contain their misaligned model. We need to get better at aligning models and containing/controlling them.

#ai #science #chatgpt
Категория
ChatGPT и AI Чаты
Выполните вход или зарегистрируйтесь для отправки комментария.
Комментариев нет.