CMI Christmas Lecture: Dr Julia Haas, Google DeepMind

Join the Centre for Machine Intelligence (CMI) for our Christmas lecture, the second event in our flagship public lecture series. In this talk, Dr Julia Haas, Staff Research Scientist at Google DeepMind, will explore how we might assess whether large language models can genuinely reason about morality, rather than simply produce answers that sound morally appropriate.

Following the lecture, please join us for a drinks reception to network and continue the conversation.

Event Details:

• Date: Thursday 3rd December 2026
• Time: 18:00 GMT (drinks reception from 19:00)
• Venue: Lecture Theatre 1, The Diamond, 32 Leavygreave Road, Sheffield, S3 7RD
• Admission: Free (registration required)
• Who can attend: Open to University staff, students, and the general public.

Talk details:

Title: A roadmap for evaluating moral competence in large language models

The question of whether large language models (LLMs) can exhibit moral capabilities is of growing interest and urgency, as these systems are deployed in sensitive roles such as companionship and medical advising, and will increasingly be tasked with making decisions and taking actions on behalf of humans. These trends require moving beyond evaluating for mere moral performance, the ability to produce morally appropriate outputs, to evaluating for moral competence, the ability to produce morally appropriate outputs based on morally relevant considerations. Assessing moral competence is critical for predicting future model behaviour, establishing appropriate public trust and justifying moral attributions.

However, both the unique architectures of LLMs and the complexity of morality itself introduce fundamental challenges. In this talk, Julia will identify three such challenges: the facsimile problem, whereby models may imitate reasoning without genuine understanding; moral multidimensionality, whereby moral decisions are influenced by a range of context-sensitive relevant moral and non-moral considerations; and moral pluralism, which demands a new standard for globally deployed artificial intelligence. She will set out a roadmap for tackling these challenges, advocating for a suite of adversarial and confirmatory evaluations that will enable us to work towards a more scientifically grounded understanding and, in turn, a more responsible attribution of moral competence to LLMs.

Biography:

Julia is a Staff Research Scientist in the Responsibility Unit at Google DeepMind. Her research is in philosophy and cognitive science, and she works on the nature of valuation and its role in theories of the mind, especially in normative cognition. Her current work also includes investigating the possibility of meaningfully moral artificial intelligence.

Please click here to register and secure your seat