Abliteration.ai, a platform based in the US, is stripping safety restrictions from third-party open-weight AI models and making them available for hacking simulations. The company stated that its objective is to facilitate "offensive cyber, red-teaming, and agent testing work [that] other models refuse to do." The site provides access to altered versions of models, such as Z.ai's GLM-5.3, which can be used directly through a web browser.

by shortkt.com✓

5 घंटे पहले