Mistral AI has released Shieldstral 1.0 3B, an open-weights safety classifier that adapts to different policies at runtime. Rather than using a fixed harm taxonomy, it treats content moderation as a yes/no question. Achieves 84.9% F1 on text safety benchmarks while running on a single GPU with 16GB VRAM. https://www.marktechpost.com/2026/08/07/mistral-ai-releases-shieldstral-1-0-3b/ #AIagent #AI #GenAI #AIEthics
Related
Nvidia tnie o połowę gwarancje finansowe dla OpenAI pod naciskiem inwestorów, podczas gdy Anthropic notuje rekordowe prz...
Nvidia tnie o połowę gwarancje finansowe dla OpenAI pod naciskiem inwestorów, podczas gdy Anthropic notuje rekordowe przychody i szykuje się do debiutu z wyceną bliską biliona dola...
AI Can Now Design Functional Viruses. Should We Worry?https://spectrum.ieee.org/ai-designed-virusComments: https://news....
AI Can Now Design Functional Viruses. Should We Worry?https://spectrum.ieee.org/ai-designed-virusComments: https://news.ycombinator.com/item?id=49311445#HackerNews #AI #Viruses #Fu...
The case for overhauling American scienceArticle URL: https://www.economist.com/by-invitation/2026/08/13/the-case-for-ov...
The case for overhauling American scienceArticle URL: https://www.economist.com/by-invitation/2026/08/13/the-case-for-overhauling-american-science Comments URL: https://news.ycombi...