Perspectives
Insights on AI security, governance, and real-world risk.
Learn how to prevent prompt injection, secure AI systems, and manage AI in production.
Follow Alice in Google → Stay updated on the latest AI security research and analysis.
Curiouser Soundbites: What a Former Google Cloud CISO Wants Leaders to Know About AI
The 5 Most Shocking LLM Weaknesses We Uncovered in 2025
Our red team researchers uncovered five unexpected LLM vulnerabilities in 2025, from hijacked reasoning to invisible tool execution. This countdown highlights the most eye-opening failures shaping AI safety today.
If I Already Do AI Pen Testing, Why Do I Need Red AI Teaming?
AI penetration testing and red teaming address different risks. This article explains why passing an AI pen test doesn’t guarantee real-world safety, and how red teaming exposes systemic weaknesses attackers actually exploit.
Be Prepared for an AI Crackdown from U.S. State Attorneys General
A coalition of 42 U.S. State Attorneys General is demanding stronger AI safeguards, audits, and accountability. This article explains what the warning means for AI developers, enterprises, and liability risk.
Disney and OpenAI’s $1 Billion Deal Hinges on AI Guardrails
Disney’s $1B partnership with OpenAI opens iconic IP to AI-generated content. This article explores why guardrails, red teaming, and real-time oversight will determine whether fan creativity can scale without risking brand trust.
Australia’s Age Gate Law and the Future of Youth Online Safety
Australia’s new age gate law sets a minimum age of 16 for social media use. This article explores why enforcement matters, why teens will still get through, and what layered safety measures are needed to truly protect young users.
Trusted by security and product teams in the world's most regulated industries
Alice brings years of adversarial intelligence expertise to AI security. We give enterprise teams the coverage that generic guardrails and one-time audits can't match.
Get a demo