Whether you are looking for an LLM with more safety guardrails or one completely without them, someone has probably built it.

Hirundo, the world's first Machine Unlearning platform for large language models (LLMs), announced measurable AI safety improvements across leading open-source models, powered by the NVIDIA technology ...

What makes this particularly dangerous in enterprise and production contexts is not just that the model gets it wrong, but ...

Boeing engineers Kevin Kwak (foreground) and Klaus Okkelberg confer with fellow team members Arvel Chappell III and Andrew Riha (both on-screen), who worked together to prototype a large language ...

Though new regulatory frameworks address fairness, accountability, and safety in AI systems, they often fail to directly mitigate the subtle communication bias in LLMs that can distort public ...

At NVIDIA GTC 2026, DeepRoute.ai presented a comprehensive introduction to its 40-billion-parameter Vision-Language-Action (VLA) Foundation Model architecture, representing a fundamental breakthrough ...

A lean team of 15 researchers, many in their twenties, at Sarvam successfully built a 105-billion-parameter foundational LLM from scratch. Spearheaded by Rahul Aralikatte, the young team managed data ...

As large language models (LLMs) gain momentum worldwide, there’s a growing need for reliable ways to measure their performance. Benchmarks that evaluate LLM outputs allow developers to track ...

In a 40-page court filing, the U.S. government argued Anthropic’s refusal to permit “all lawful uses” of Claude made the company too risky for national security systems.

As new large language models, or LLMs, are rapidly developed and deployed, existing methods for evaluating their safety and discovering potential vulnerabilities quickly become outdated. To identify ...

The Uncertainty Engine is guiding research in fusion plasma physics. Could similar approaches benefit fission research as well?