Podcast
AI Security Revolution: SplxAI Founders Expose Chatbot Threats
Discover AI security insights: SplxAI's platform, prompt injection risks, red teaming, DevSecOps integration, and proactive measures for secure AI applications.
Marius Poskus
Ante Gojsalić
DATE
May 17, 2024
TIME & LENGTH
45 min
STATUS
Available on demand
LANGUAGE
English
The 11th episode of the Cyber Diaries Podcast features Marius Poskus, CEO at MP Cybersecurity Services and Ante Gojsalić, Co-Founder and CTO at SplxAI and dives into SplxAI’s innovative solutions for AI security, addressing critical challenges like prompt injection, context leakage, and hallucinations. Ante shares his journey from AI development to cybersecurity and discusses how tools like the SplxAI Platform enable proactive risk management through automated scanning and red teaming integrated into DevSecOps pipelines. Key insights include fostering collaboration between AI engineers and security teams, adapting to advancements like multimodal inputs, and leveraging AI-powered risk mapping.
Securing AI Applications: Tackling Prompt Injection, Context Leakage, and Evolving Risks
Proactive Risk Management: Tools like Probe integrate red teaming into DevSecOps pipelines, enabling continuous and automated vulnerability assessments.
Key Challenges in AI Security: Prompt injection, context leakage, and hallucinations are highlighted as top threats to conversational AI applications.
Future-Ready Solutions: SplxAI focuses on AI-powered risk mapping, multimodal security capabilities, and fostering collaboration between engineers and security teams.
WhitePaper
