Last Week in AI cover image

#200 - ChatGPT Roadmap, Musk OpenAI Bid, Model Tampering

Last Week in AI

00:00

AI Utility Engineering and Value Systems

This chapter investigates a research paper on 'Utility Engineering,' focusing on the emergent value systems in AI during training. It highlights biases in AI models like GPT-40, examining how utility functions influence decision-making and the implications of training data. Additionally, it discusses the release of a risk management framework by XAI aimed at assessing potential AI-related threats, emphasizing the importance of transparency and accountability.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app