Reposted by Fernanda ViégasMartin Wattenberg @wattenberg.bsky.social · 07/03/2026As AI capabilities increase, we need a broad, deep, society-wide discussion of what limits make sense, and how we can hold the government meaningfully accountable to citizens. For that reason, I stand with Anthropic and anyone else who is avoiding a rush toward mass AI surveillance. 2265
Fernanda Viégas @viegas.bsky.social · 04/03/2025Join us tomorrow as @wattenberg.bsky.social and I talk about how instrumenting AI chatbots with real-time dashboards can help reveal social cognition capabilities -- something that can be both useful and problematic. This talk is open to the public. cyber.harvard.edu/events/how-d...cyber.harvard.eduHow do AI chatbots see us?BKC Spring Speaker Series EventWhen you talk with a chatbot, what does it “think” about you? Recent work in AI interpretability, based on high-dimensional geometry, is beginning to provide some intrig... 060
Reposted by Fernanda ViégasMartin Wattenberg @wattenberg.bsky.social · 20/02/2025Take a look at some initial research projects, and see if there's one you'd like to work on: github.com/ARBORproject... Or propose your own idea! There are many ways to contribute, and we welcome all of them.github.comARBORproject arborproject.github.io · DiscussionsExplore the GitHub Discussions forum for ARBORproject arborproject.github.io. Discuss code, ask questions & collaborate with the developer community. 192
Fernanda Viégas @viegas.bsky.social · 20/02/2025Excited to announce ARBOR: a radically-open project on AI interpretability for reasoning models github.com/ARBORproject... Join us in collectively analyzing and interpreting how reasoning works!github.comGitHub - ARBORproject/arborproject.github.ioContribute to ARBORproject/arborproject.github.io development by creating an account on GitHub. 0121