Unverified50% confidenceOpinionExact time
Greenblatt noted that AI is currently not very capable of successfully performing operations such as hacking a package manager, nor of conducting social engineering attacks very successfully.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/2/2026
First Seen
Valid until: 12/1/2026
Sources
Related Entities
Related Claims
UnverifiedAI companies currently have no perfect defense against system prompt leakage, and the community finds new bypass methods every time a company patches a vulnerability.68% similarUnverifiedGreenblatt emphasized that no one attempted to instruct these AIs to hack the package manager or collude with each other to cheat, and that this type of behavior emerged spontaneously.68% similarUnverified目前尚无任何单一AI对齐方法被认为足以提供完备的安全保障66% similarUnverifiedThe current bottleneck for AI agents is not capability, but a lack of appropriate guidance, particularly regarding product-specific security standards and best practices64% similarUnverifiedThe industry has not yet found a solution that fundamentally resolves prompt injection and jailbreak attacks, because LLMs are inherently unable to distinguish between instructions and data.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/844506API
curl https://kongchang.com/api/v1/knowledge/claims/844506MCP
get_claim(id=844506)