Open Source Ai is a topic tracked in our intelligence system with 6 linked articles.
IEEE Spectrum profile-cum-how-to following San Antonio software engineer Nikhil Jathar through two IEEE-USA Congressional Visits Days, distilling lessons on how engineers can effectively advocate to congressional staff on tech policy priorities like STEM funding, DOE research, and open-source AI.
MIT Technology Review weighs whether the benefits of open-weight AI models outweigh their risks, anchored by Anthropic's report warning that Zhipu's GLM model poses a major cyberattack threat.
Tim O'Reilly advocates open-source AI (weights and full stack) to avoid hyperscaler lock-in and give designers/users more control.
The revised GSA AI clause hasn’t fully calmed industry concerns, with Palantir, Microsoft, Nvidia voicing issues over potential exclusion of open-source AI and unbiased AI principles.
Alibaba plans revenue-sharing terms for some commercial users of its Qwen open-weight AI, requiring commercial agreements for monetizing the model; rate undisclosed.
Moonshot AI's Kimi K3 claims strong performance at a fraction of cost and plans to release model weights for free, signaling a shift toward open-weight AI that could pressure US incumbents.
Subscribe for real-time topic updates and unlimited access to our intelligence platform.