Injection(前回01位) • LLM02:2025 Sensitive Information Disclosure(前回06位) • LLM03:2025 Supply Chain(前回05位) • LLM04:2025 Data and Model Poisoning(前回03位) • LLM05:2025 Improper Output Handling(前回02位) • LLM06:2025 Excessive Agency (前回08位) • LLM07:2025 System Prompt Leakage(新登場) • LLM08:2025 Vector and Embedding Weaknesses(新登場) • LLM09:2025 Misinformation(前回09位) • LLM10:2025 Unbounded Consumption(前回04位)