Hi Fellows We had a bug so you couldn't get into your dashboard Thanks for our friends who reported this we fixed the bug. If it still persisting, try to hard reload to clear the cache If you encounter more issues, please let us know

New from our STEM Fellows Program: Designing more efficient quantum circuits with Claude. Given plain-language instructions, Claude Mythos 5 designed quantum circuits that improve on the best published ones, sometimes by an order of magnitude or more. More efficient circuits reduce the resources needed by quantum algorithms, which could bring quantum computing applications in chemistry and materials science within reach sooner. Read more:

What is the Flagship Fellowship? Discover a 12-week, paid summer program that pushes fellows to stretch beyond existing scholarship and embrace uncertainty while working side-by-side with our origination teams. Fellows use Flagship’s company-creation process and AI tools to generate and refine ideas that have the potential to become breakthrough companies. Learn more about the Fellowship at an upcoming info session: 📅 October 15, 10–11 AM EDT 📅 November 10, 1–2 PM EDT 📅 December 7, 11 AM–12 PM EDT Sign up here: #FlagshipFellowship# #WeAreFlagship#

📢 Calling future Fellows! 👉🏽 Who you are: A university undergrad in the venture/entrepreneurship world, or with a background in engineering, product, or design 👉🏽 What makes you, you: An independent thinker and team player — curious, engaged, ready to dive in 👉🏽 What you'll get: ✅ Hands-on experience in the business of VC ✅ Direct collaboration with investment professionals ✅ The chance to build your own investment thesis ✅ Deeper ties to your campus entrepreneurial ecosystem ✅ Cool swag (yes, really) Applications close October 16, 2026 — don't wait ⏳ ✍🏽 Apply now:

New Anthropic Fellows research: Model Spec Midtraining (MSM). Standard alignment methods train AIs on examples of desired behavior. But this can fail to generalize to new situations. MSM addresses this by first teaching AIs how we would like them to generalize and why.

In new Anthropic Fellows research, we discuss “introspection adapters": a tool that allows language models to self-report behaviors they've learned during training—including potential misalignment.