相关论文: The Alignment Trap: Complexity Barriers

Safety alignment is an important procedure before the official deployment of a Large Language Model (LLM). While safety alignment has been extensively studied for LLM, there is still a large research gap for Large Reasoning Models (LRMs)…

As AI becomes more capable, we entrust it with more general and consequential tasks. The risks from failure grow more severe with increasing task scope. It is therefore important to understand how extremely capable AI models will fail: Will…

Large language models (LLMs) have become increasingly sophisticated, leading to widespread deployment in sensitive applications where safety and reliability are paramount. However, LLMs have inherent risks accompanying them, including bias,…

Sycophancy (overly agreeable or flattering behavior) poses a fundamental challenge for human-AI collaboration, particularly in high-stakes decision-making domains such as health, law, and education. A central difficulty in studying…

With AI systems becoming more powerful and pervasive, there is increasing debate about keeping their actions aligned with the broader goals and needs of humanity. This multi-disciplinary and multi-stakeholder debate must resolve many…

The success of modern Artificial Intelligence (AI) technologies depends critically on the ability to learn non-linear functional dependencies from large, high dimensional data sets. Despite recent high-profile successes, empirical evidence…

The rapid trend of deploying artificial intelligence (AI) and machine learning (ML) systems in socially consequential domains has raised growing concerns about their trustworthiness, including potential discriminatory behaviours. Research…

The burgeoning integration of artificial intelligence (AI) into human society brings forth significant implications for societal governance and safety. While considerable strides have been made in addressing AI alignment challenges,…

The guiding principle of AI alignment is to train large language models (LLMs) to be harmless, helpful, and honest (HHH). At the same time, there are mounting concerns that LLMs exhibit a left-wing political bias. Yet, the commitment to AI…

The rise of general-purpose artificial intelligence (AI) systems, particularly large language models (LLMs), has raised pressing moral questions about how to reduce bias and ensure fairness at scale. Researchers have documented a sort of…

Artificial agents now generate behavior rich enough to invite trust, surprise, and concern, yet our evaluation tools still privilege capability scores over psychological structure. This paper argues that the philosophical impasse between…

Trustworthy capability evaluations are crucial for ensuring the safety of AI systems, and are becoming a key component of AI regulation. However, the developers of an AI system, or the AI system itself, may have incentives for evaluations…

Foundation models excel in stable environments, yet often fail where reliability matters most: medicine, finance, and policy. This Fidelity Paradox is not just a data problem; it is structural. In domains where rules change over time, extra…

When AI interacts with the physical world -- as a robot or an assistive agent -- new safety challenges emerge beyond those of purely ``digital AI". In such interactions, the potential for physical harm is direct and immediate. How well do…

Optimizing a given metric is a central aspect of most current AI approaches, yet overemphasizing metrics leads to manipulation, gaming, a myopic focus on short-term goals, and other unexpected negative consequences. This poses a fundamental…

There is growing evidence that independently trained AI systems come to represent the world in the same way. In other words, independently trained embeddings from text, vision, audio, and neural signals share an underlying geometry. We call…

The linear programming (LP) approach is, together with value iteration and policy iteration, one of the three fundamental methods to solve optimal control problems in a dynamic programming setting. Despite its simple formulation,…

The rapid integration of Large Language Models (LLMs) into educational assessment rests on the unverified assumption that instruction following capability translates directly to objective adjudication. We demonstrate that this assumption is…

As AI systems become more advanced, companies and regulators will make difficult decisions about whether it is safe to train and deploy them. To prepare for these decisions, we investigate how developers could make a 'safety case,' which is…

The remarkable capabilities of Large Language Models (LLMs) make them increasingly compelling for adoption in real-world healthcare applications. However, the risks associated with using LLMs in medical applications have not been…