This paper argues that a range of current AI systems have learned how to deceive humans.
We define deception as the systematic inducement of false beliefs in the pursuit of some …

An oracle that correctly predicts the outcome of every particle physics experiment, the
products of every possible chemical reaction or the function of every protein would …

In reinforcement learning from human feedback, it is common to optimize against a reward
model trained to predict human preferences. Because the reward model is an imperfect …

Reinforcement learning algorithms typically struggle in the absence of a dense, well-shaped
reward function. Intrinsically motivated exploration methods address this limitation by …

This work identifies 18 foundational challenges in assuring the alignment and safety of large
language models (LLMs). These challenges are organized into three different categories …

Machine learning (ML) systems are rapidly increasing in size, are acquiring new
capabilities, and are increasingly deployed in high-stakes settings. As with other powerful …

Deep learning has triggered the current rise of artificial intelligence and is the workhorse of
today's machine intelligence. Numerous success stories have rapidly spread all over …

AI alignment aims to make AI systems behave in line with human intentions and values. As
AI systems grow more capable, the potential large-scale risks associated with misaligned AI …

In coming decades, artificial general intelligence (AGI) may surpass human capabilities at
many critical tasks. We argue that, without substantial effort to prevent it, AGIs could learn to …

Through multi-agent competition, the simple objective of hide-and-seek, and standard
reinforcement learning algorithms at scale, we find that agents create a self-supervised …