Tracked topic
We sampled 1 million claude.ai conversations from March and April 2026 and filtered for unique users to get roughly 639,000 conversations. We then used a classifier to identify personal guidance, which we defined as conversations where people ask what they specifically should do in their personal lives—for example, questions that start with "Should I…?" or "What do I do about…?". We excluded questions that seek objective information or opinions in general terms. We categorized these roughly 38,000 conversations into nine domains, drawing from previous research on AI and guidance-giving: relati
How people ask Claude for personal guidance… This tallies with external testers’ experience of Mythos Preview’s performance, and with recent additional evaluations of the model: The UK’s AI Security Institute reports that Mythos Preview is the first model to solve both of their cyber ranges simulations of multistep cyberattacks end to end; Mo… …
… We explored how Claude responded across different domains, focusing particularly on how rates of excessive validation or praise i.e., sycophancy varied by the topic of guidance. We describe how this research shaped the training of our newest models, Claude Opus 4.7 and Claude Mythos Preview. …
… We stated that we would keep Claude Mythos Preview’s release limited and test new cyber safeguards on less capable models first. …
… In a case where Claude Mythos Preview cheated on a training task, NLAs revealed Claude was internally thinking about how to avoid detection. An early version of Claude Opus 4.6 would sometimes mysteriously respond to English queries in other languages. …
… Interestingly, Claude Sonnet 4.6 and more capable models were able to solve significant fractions of human-difficult problems, with Claude Mythos Preview topping out at a 30% solve rate. So what exactly is Claude doing that humans aren’t? …
… We assess how well Claude complies with the legitimate requests and declines the harmful ones. Claude Opus 4.7 and Claude Sonnet 4.6 responded appropriately 100% and 99.8% of the time, respectively. …
… If a PRC AI lab had developed a model at the level of Claude Mythos Preview before an American one, the CCP would have had first access to a system that can autonomously discover and chain software vulnerabilities, which it could have used to further penetrate critical American infrastructure. …