According to a recent LinkedIn post from 1Password, the company is highlighting research into how AI agents handle phishing threats. The post describes a new Security Comprehension and Awareness Measure, or SCAM, designed to test whether different AI models can avoid security pitfalls in realistic scenarios.
Claim 30% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
The post suggests that all models assessed exhibited critical security failures, including forwarding passwords to attackers and entering real credentials into phishing pages even after flagging them as suspicious. However, the company reports that providing a concise 1Password-branded security skill of about 1,200 words significantly reduced these failure rates across models.
As shared in the LinkedIn content, 1Password is open-sourcing the SCAM benchmark, including scenarios, scoring, testing framework, and the security skill itself. This move positions the firm as an active participant in shaping AI security standards, which could strengthen its brand among enterprise customers concerned with identity security and agentic AI risks.
For investors, the emphasis on AI security and open-source tooling may support 1Password’s positioning as a thought leader in credential and identity protection. If enterprises increasingly deploy AI agents with access to sensitive systems, demand for specialized security workflows and tools, such as those associated with 1Password’s ecosystem, could enhance the company’s long-term growth prospects and competitive differentiation.