English
Related papers

Related papers: An alignment safety case sketch based on debate

200 papers

To make AI systems broadly useful for challenging real-world tasks, we need them to learn complex human goals and preferences. One approach to specifying complex goals asks humans to judge during training which agent behaviors are safe and…

Machine Learning · Statistics 2018-10-23 Geoffrey Irving , Paul Christiano , Dario Amodei

The emergence of pre-trained AI systems with powerful capabilities across a diverse and ever-increasing set of complex domains has raised a critical challenge for AI safety as tasks can become too complicated for humans to judge directly.…

Artificial Intelligence · Computer Science 2023-11-27 Jonah Brown-Cohen , Geoffrey Irving , Georgios Piliouras

We sketch how developers of frontier AI systems could construct a structured rationale -- a 'safety case' -- that an AI system is unlikely to cause catastrophic outcomes through scheming. Scheming is a potential threat model where AI…

For some problems, humans may not be able to accurately judge the goodness of AI-proposed solutions. Irving et al. (2018) propose that in such cases, we may use a debate between two AI systems to amplify the problem-solving capabilities of…

Artificial Intelligence · Computer Science 2021-03-17 Vojtěch Kovařík , Ryan Carey

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…

Computers and Society · Computer Science 2024-03-20 Joshua Clymer , Nick Gabrieli , David Krueger , Thomas Larsen

As AI grows more powerful, it will increasingly shape how we understand the world. But with this influence comes the risk of amplifying misinformation and deepening social divides-especially on consequential topics where factual accuracy…

A core challenge in the development of increasingly capable AI systems is to make them safe and reliable by ensuring their behaviour is consistent with human values. This challenge, known as the alignment problem, does not merely apply to…

Machine Learning · Computer Science 2023-11-07 Raphaël Millière

The field of AI alignment aims to steer AI systems toward human goals, preferences, and ethical principles. Its contributions have been instrumental for improving the output quality, safety, and trustworthiness of today's AI models. This…

Artificial Intelligence · Computer Science 2024-11-26 Robert West , Roland Aydin

This paper contributes to the nascent debate around safety cases for frontier AI systems. Safety cases are structured, defensible arguments that a system is acceptably safe to deploy in a given context. Historically, they have been used in…

Computers and Society · Computer Science 2026-03-11 Shaun Feakins , Ibrahim Habli , Phillip Morgan

AI alignment is about ensuring AI systems only pursue goals and activities that are beneficial to humans. Most of the current approach to AI alignment is to learn what humans value from their behavioural data. This paper proposes a…

Artificial Intelligence · Computer Science 2023-10-06 Pei-Yu Chen , Myrthe L. Tielman , Dirk K. J. Heylen , Catholijn M. Jonker , M. Birna van Riemsdijk

Frontier AI systems are rapidly advancing in their capabilities to persuade, deceive, and influence human behaviour, with current models already demonstrating human-level persuasion and strategic deception in specific contexts. Humans are…

Artificial Intelligence · Computer Science 2025-07-18 Rishane Dassanayake , Mario Demetroudi , James Walpole , Lindley Lentati , Jason R. Brown , Edward James Young

We present our Balanced, Integrated and Grounded (BIG) argument for assuring the safety of AI systems. The BIG argument adopts a whole-system approach to constructing a safety case for AI systems of varying capability, autonomy and…

Computers and Society · Computer Science 2025-04-01 Ibrahim Habli , Richard Hawkins , Colin Paterson , Philippa Ryan , Yan Jia , Mark Sujan , John McDermid

Existing evaluations of AI misuse safeguards provide a patchwork of evidence that is often difficult to connect to real-world decisions. To bridge this gap, we describe an end-to-end argument (a "safety case") that misuse safeguards reduce…

Machine Learning · Computer Science 2025-05-26 Joshua Clymer , Jonah Weinbaum , Robert Kirk , Kimberly Mai , Selena Zhang , Xander Davies

AI safety via debate and reinforcement learning from AI feedback (RLAIF) are both proposed methods for scalable oversight of advanced AI systems, yet no formal framework relates them or characterizes when debate offers an advantage. We…

Machine Learning · Computer Science 2026-03-06 Robin Young

As AI systems are used to answer more difficult questions and potentially help create new knowledge, judging the truthfulness of their outputs becomes more difficult and more important. How can we supervise unreliable experts, which have…

Artificial Intelligence · Computer Science 2023-11-16 Julian Michael , Salsabila Mahdi , David Rein , Jackson Petty , Julien Dirani , Vishakh Padmakumar , Samuel R. Bowman

While multi-agent debate has been proposed as a promising strategy for improving AI reasoning ability, we find that debate can sometimes be harmful rather than helpful. Prior work has primarily focused on debates within homogeneous groups…

Computation and Language · Computer Science 2025-10-14 Andrea Wynn , Harsh Satija , Gillian Hadfield

The core premise of AI debate as a scalable oversight technique is that it is harder to lie convincingly than to refute a lie, enabling the judge to identify the correct position. Yet, existing debate experiments have relied on datasets…

In this paper we discuss how systems with Artificial Intelligence (AI) can undergo safety assessment. This is relevant, if AI is used in safety related applications. Taking a deeper look into AI models, we show, that many models of…

Artificial Intelligence · Computer Science 2021-05-17 Jens Braband , Hendrik Schäbe

Alignment of artificial intelligence (AI) encompasses the normative problem of specifying how AI systems should act and the technical problem of ensuring AI systems comply with those specifications. To date, AI alignment has generally…

While autonomous agents often surpass humans in their ability to handle vast and complex data, their potential misalignment (i.e., lack of transparency regarding their true objective) has thus far hindered their use in critical applications…

Artificial Intelligence · Computer Science 2024-12-03 Frédéric Berdoz , Roger Wattenhofer
‹ Prev 1 2 3 10 Next ›