48% report needing to get better at defending against AI-powered cyber adversaries.

59% of executives say that it is becoming more difficult for employees to identify real threats as AI-powered technologies make attacks more sophisticated.

46% of respondents say their organizations use AI/ML to prevent cyberattacks.

84% recognise the role of the AppSec leader as more important now than ever. More than 84% believe their role is more important now than it was a few years ago. This increased importance is linked to factors such as growing challenges from AI-generated code and open source software.

When prompted to generate secure code, GPT-4o still produced insecure outputs vulnerable to 8 out of 10 issues.

25% of reported security issues in Agentic AI remain open.

68% of organizations surveyed have experienced data leakage incidents specifically related to employees sharing sensitive information with AI tools.

OpenAI’s GPT-4o had the lowest performance, scoring a 1/10 secure code result using "naive" prompts.

In January 2025, Skyhigh Security recorded DeepSeek usage by 43% of customers.

11% of files uploaded to AI applications include sensitive corporate content.

In January 2025, customers uploaded a combined 176GB of data into the DeepSeek AI chatbot.

In the UK, security risks associated with third-party suppliers have surged by "more than ten percentage points" as a top concern, largely driven by the integration of third-party AI solutions

Some open security issues in Agentic AI are lingering for 1,200-plus days.

83% of AI applications don’t support integration with multi-factor authentication (MFA) tools

Over 700 issues in Agentic AI repositories remain unaddressed.

In response to simple, “naive” prompts, all LLMs tested generated insecure code vulnerable to at least 4 of the 10 common CWEs.

94% of all AI services are at risk for at least one of the top Large Language Model (LLM) risk vectors, including prompt injection/jailbreak, malware generation, toxicity, and bias.

60% of top vulnerabilities found in Agentic AIwere access control-related

44% of security leaders surveyed plan to prioritize security infrastructure oversight and implementation, much of which now focuses on securing AI systems and preventing data leakage.

Skyhigh Security research reveals a 200% increase in AI application traffic within the last year. This compares to a 23% increase in traffic to non-AI applications.