A Summary of Recent Work (July 2026)
AGI Safety & Alignment Research at GDM
SubscribeSign in
A Summary of Recent Work (July 2026)<br>By Rohin Shah and Seb Farquhar.
GDM Alignment Research Blog<br>Jul 31, 2026
16
Share
It’s been nearly two years since our last major update in August 2024 and we wanted to share another recap of our recent work. Things have changed a lot since then. We are now fully in the midgame, and focus more on landing things in production.<br>Who are we?
We are the AGI Safety and Alignment Team (ASAT), the main group at Google DeepMind working directly on technical approaches to existential risk from AI systems. Last year we published An Approach to Technical AGI Safety and Security, which remains the best place to read our overarching vision.<br>Highlights
Norms around chain of thought. Our impression is that our work meaningfully moved the field away from beliefs along the lines of “chain of thought is often unfaithful and so not worth using” towards beliefs along the lines of “chain of thought is a very useful tool that is worth preserving”, leading to a tentative industry consensus on its importance. We have also published substantial technical research that enables companies to preserve chain of thought transparency for longer than would have happened by default.<br>We think this is a big deal: extending the period where model reasoning is relatively transparent enables better science on more powerful AI systems, better model forensics on future warning shots, and stronger bootstrapping of control monitors.<br>Frontier Safety. We substantially strengthened the Frontier Safety Framework (FSF), and were the first company to introduce a section on misalignment in such a framework. The scope of Frontier Safety has grown a lot over the past two years, and is now a cross-functional effort including many teams across Google. We believe the FSF played a key role in maintaining Google’s situational awareness around severe risks, and particularly around starting to develop mitigations well in advance of when they would be needed. For example, this was particularly useful for landing probes, which are quite a complex change in production as they touch many parts of the infra stack.<br>Technical planning. We published An Approach to Technical AGI Safety and Security as well as the GDM AI Control Roadmap, which go into substantial detail about the specific areas of work we believe are best to focus on in AGI safety and control respectively. We have found it very useful for organizing our work internally, and ensuring that our work makes progress towards the right goals.<br>Agent Control & Monitorability
While monitoring has formed a part of our thinking for a long time, a couple of years ago it became a priority to do empirical work on it, since it seemed plausible that it would soon be needed in practice.<br>At the time, the prevailing opinion we heard was that the chain of thought was unfaithful (and therefore, by implication, effectively useless). This seemed wrong to us: while the chain of thought is frequently unfaithful, this tends to be on easy tasks where the chain of thought isn’t load bearing to get the correct answer. However, this doesn’t apply to difficult tasks requiring significant reasoning, as would be the case for many of the most concerning misalignment threat models. In such cases, the chain of thought is necessary to complete the task correctly, and so we should expect it to be informative.<br>Along with a variety of other organizations, we coauthored the position paper Chain-of-Thought Monitorability: A New and Fragile Opportunity for AI Safety, in which the necessity argument is placed front and center in Figure 1. We validated the necessity argument empirically in the paper When Chain of Thought is Necessary, Language Models Struggle to Evade Monitors, and formalized the concept of Opaque Serial Depth as a theoretically principled metric that can quantify the extent to which the necessity argument still applies for new model architectures. We then applied these ideas to text diffusion models in the paper How Transparent is DiffusionGemma?, with the hope of setting standard practices for assessing the impacts of latent reasoning architectures on model transparency.<br>The necessity argument only shows that relevant information must exist in the chain of thought; it does not say anything about whether we can understand that information. In practice, due to the natural language prior learned from pretraining, LLMs tend to think in natural language that we can understand. We built a simple pragmatic autorater to measure this default legibility. We also developed a framework to analyze when we can safely optimize chain of thought without significantly threatening the default legibility of chain of thought.<br>To tie these arguments more closely to misalignment risks, we developed capability evaluations for stealth and situational awareness. These evaluations serve two purposes: (1) assessing...