Unverified50% confidenceOpinionExact time
Ghodsi stated that AI models are very good at breaching defenses and that enterprises must invest heavily in cybersecurity to protect software systems.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/24/2026
First Seen
Valid until: 12/23/2026
Sources
Related Entities
Related Claims
UnverifiedAI驱动的网络安全工具已从概念验证阶段进入大规模部署阶段74% similarVerifiedAI安全正在从产品的可选附加项变成基础设施级的必选项71% similarUnverifiedCybersecurity is evolving from protecting people and networks toward managing and controlling AI agents.71% similarUnverifiedModel weights are among the most core assets of frontier AI companies, and a gap in security protection could expose them to risks of theft or misuse70% similarVerifiedAnthropic has consistently regarded AI safety as the company's core mission70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/947695API
curl https://kongchang.com/api/v1/knowledge/claims/947695MCP
get_claim(id=947695)