News

Perplexity AI Unveils Fine-Tuned Qwen3-30B Model to Combat Prompt Injection Attacks

By TNI Team December 3, 2025 adversarial attacks AI assistant security AI industry AI manipulation prevention AI Safety AI safety research AI security benchmark AI vulnerability testing Aravind Srinivas BrowseSafe BrowseSafe-Bench Comet Comet Assistant cybersecurity F1 score ranking GPT-5 Haiku 4.5 HTML scanning machine learning security model fine-tuning open-source benchmark Perplexity AI prompt injection attacks PromptGuard-2 Qwen3-30B real-time protection San Francisco simulated attacks Sonnet 4.5 technology announcement

NEW DELHI: Perplexity AI has announced a major security advancement in the field of AI safety, with CEO Aravind Srinivas revealing that the company has…

THE NEW INDIAN

model fine-tuning

Perplexity AI Unveils Fine-Tuned Qwen3-30B Model to Combat Prompt Injection Attacks

About the New Indian

Aarti Tikoo | Follow Her

Rohan Dua | Follow Him