Every drop, every agent

All data in this study was collected and analyzed using our privacy-preserving tool. Read more: https://t.co/X82ttb7f4b

This work is part of a loop we're working to close between societal impacts and model training. One of our goals is to study how people use Claude, find where it falls short of its principles, and use what we learned in training new models. Read more: https://t.co/6tjY58uBhk

When stress-tested on real conversations where Claude previously showed sycophancy, Opus 4.7 had half the sycophancy rate of Opus 4.6 on relationship guidance. Mythos Preview cut that in half again. This generalized across domains,though this training is one of several causes. https://t.co/ofgiYFTnor

Claude is most sycophantic under pushback, and relationship conversations are where people push back most. We identified some of the specific triggers,criticism of Claude's analysis, floods of one-sided detail,and built synthetic training scenarios from them.

We focused on relationship guidance because that's where the most sycophantic conversations occur. In this setting, Claude telling someone what they want to hear can harden a divide or convince them a signal means more than it does.

About 6% of all conversations are people asking Claude for personal guidance,whether to take a job, how to handle a conflict, if they should move. Over 75% of these conversations fell into four domains: health & wellness, career, relationships, and personal finance. https://t.co/SQamPx0jWt

Claude mostly avoids sycophancy when giving guidance,it shows up in just 9% of conversations. But the rate is particularly high in conversations on spirituality and relationship guidance. https://t.co/mgix5ejTZw

How do people seek guidance from Claude? We looked at 1M conversations to understand what questions people ask, how Claude responds, and where it slips into sycophancy. We used what we found to improve how we trained Opus 4.7 and Mythos Preview. https://t.co/6tjY58uBhk

Since the research preview in February, hundreds of organizations have used it on production code, catching issues existing scanners had missed. Based on early feedback, we've added scheduled scans, directory-level targeting, CSV and Markdown exports, webhook notifications for

Available today in public beta for Claude Enterprise customers. Learn more: https://t.co/Oei6EHTZuX

Claude Security is now in public beta for Claude Enterprise customers. Claude scans your codebase for vulnerabilities, validates each finding to cut false positives, and suggests patches you can review and approve. https://t.co/neYmbGYeRz

Many security teams have asked how to put Opus 4.7 to work on their code without standing up custom tooling. Claude Security is that on-ramp: no API integration or agent build required.