Unverified50% confidenceFactExact time
不同AI模型对特定编程场景倾向于幻觉的包名存在跨模型的一致性规律，使攻击者只需少量查询即可建立高价值目标包名列表
1
Sources
50%
Confidence
Long-term
Relevance
7/10/2026
First Seen
Sources
AI Coding Security: A CLI Tool Designed to Block Vulnerable Dependencies for Agents
hackernewshackernews7/1/2026
Related Claims
Unverified研究表明AI模型在特定场景下会重复生成相同的幻觉包名，意味着攻击者可以系统性地收集这些名称85% similarUnverifiedAutomated jailbreaking uses another AI model to automatically generate jailbreak prompts targeting a victim model, based on the principle of adversarial prompt search.79% similarUnverifiedAI Agent模式下模型会生成大量用户不可见的中间推理步骤，这些步骤同样计入Token消耗78% similarUnverifiedThe routing layer in AI inference scenarios typically uses lightweight classifiers (such as DistilBERT or in-house small models) to score incoming requests in real time.78% similarUnverifiedChoosing an AI model should be based on one's own core use case rather than blindly following comprehensive leaderboard rankings.77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/426313API
curl https://kongchang.com/api/v1/knowledge/claims/426313MCP
get_claim(id=426313)