Security News > 2024 > August > The AI balancing act: Unlocking potential, dealing with security issues, complexity
The rapid integration of AI and GenAI technologies creates a complex mix of challenges and opportunities for organizations.
The most urgent security risks for GenAI users are all data-related.
The most worrying AI threats include GenAI model prompt hacking, Large Language Model data poisoning, Ransomware as a Service, GenAI processing chip attacks, Application Programming Interface breaches, and GenAI phishing.
41% say GenAI has the most potential to address cyber alert fatigue.
92% of security pros have security concerns around generative AI, with specific apprehensions including employees entering sensitive company data into an AI tool, using AI systems trained with incorrect or malicious data, and falling for AI-enhanced phishing attempts.
As today's risks are increasingly driven by AI and GenAI, the way employees work, and the proliferation of cloud applications, respondents state they need more visibility into source code sent to repositories, files sent to personal cloud accounts, and customer relationship management system data downloads.
News URL
https://www.helpnetsecurity.com/2024/08/15/ai-genai-security-risks/
Related news
- AI accelerates code development faster than security teams can keep up (source)
- Embedding AI security from the get go (source)
- AI security 2024: Key insights for staying ahead of threats (source)
- Unlock the Future of Cybersecurity: Exclusive, Next Era AI Insights and Cutting-Edge Training at SANS Network Security 2024 (source)
- AI for application security: Balancing automation with human oversight (source)