Adversarial examples can force computer-use artificial intelligence (AI) agents to execute arbitrary code. To aid AI researchers in evaluating robustness of agentic models, Intel Labs researchers open sourced an adversarial image injection proof of concept (PoC) against computer-use AI agents such as UI-TARS
-
-
Articles récents
- Rethinking AI Infrastructure: How NetApp and Intel Are Unlocking the Future with AIPod Mini
- Intel Labs Open Sources Adversarial Image Injection to Evaluate Risks in Computer-Use AI Agents
- Optimizing LLM Inference on Intel® Gaudi® Accelerators with llm-d Decoupling
- Robots Meet Humans: Intel Labs Extends Robotics Safety to Cover 3D Environments
- Bringing AI Back to the Device: Real-World Transformer Models on Intel® AI PCs
-
Neural networks news
Intel NN News
- Rethinking AI Infrastructure: How NetApp and Intel Are Unlocking the Future with AIPod Mini
In an era dominated by the narrative that “AI equals GPUs,” a quiet revolution is […]
- Intel Labs Open Sources Adversarial Image Injection to Evaluate Risks in Computer-Use AI Agents
Adversarial examples can force computer-use artificial intelligence (AI) agents to execute […]
- Optimizing LLM Inference on Intel® Gaudi® Accelerators with llm-d Decoupling
Discover how Intel® Gaudi® accelerators and the llm-d stack improve large language model […]
- Rethinking AI Infrastructure: How NetApp and Intel Are Unlocking the Future with AIPod Mini
-