Framework for frontier AI and dawning of a new age, bioresilience, optimizing against safety representations
A Framework for Frontier AI and the Dawning of a New Age
“This is a pivotal moment in human history. Artificial General Intelligence (AGI), a system that exhibits all the cognitive capabilities the brain has, is probably only a few short years away. When we look back on this time in the decades to come, I think we will realise we were standing in the foothills of the singularity - nothing less than the dawning of a new age for humanity…
The rapid progress we’re seeing in AI requires a new approach to testing frontier AI model capabilities that is dynamic, adaptable, and rigorous. The US is well positioned, given its economic and technical standing, to take the first step in developing such a framework. It could establish a new Standards Body modelled on a federally overseen public-private partnership or self-regulatory organisation, much like the Financial Industry Regulatory Authority (FINRA), with a board that includes independent leading technical experts and open-source representatives. Funding would need to be substantial and likely mostly come from industry, in order to attract world-class technical talent and provide the necessary compute resources for large-scale testing.”
Google DeepMind’s Approach to Bioresilience
“The global biosecurity landscape is rapidly evolving. Shifting natural ecosystems, global travel and the potential misuse of AI require greater vigilance — yet AI is also a critical tool for our response. We need frontier AI models, and the scientific advances they will enable, to respond to these challenges and help make society more resilient to events like future outbreaks.”
“Behavioral alignment in large language models often masks fragile internal safety representations. Recent work suggests that refusal behavior is mediated by low-dimensional directions in activation space. This raises questions about how such representations are structured, localized, and accessed by optimization. We study adversarial suffix attacks as a probe of representational alignment. We introduce Activation-Guided GCG, which replaces output-based objectives with losses that directly target a model’s internal refusal direction. Across several objective variants, we find that suppressing refusal globally across all layers and positions is more effective than targeting a single layer-position pair. This suggests that safety representations are distributed across the forward pass rather than causally localized to a single site. We further introduce Soft-GCG, a continuous relaxation of discrete suffix optimization using Gumbel-Softmax. Soft-GCG achieves a 33 × speedup over standard GCG while improving attack success rates. Evaluating across model scales, we find that smaller models remain vulnerable while larger models resist both activation- and suffix-based attacks at our compute-constrained settings, consistent with larger and better safety trained models being harder to jailbreak. Together, our results clarify how safety mechanisms are encoded and can be broken in contemporary models. These insights provide concrete guidance for designing more robust and representation-aware alignment strategies.”