PwC identifies AI attacks as largest cyber readiness gap

PwC's latest survey reveals that attacks targeting artificial intelligence systems have emerged as the most significant cybersecurity preparedness gap for organizations worldwide. Half of security leaders surveyed placed AI system attacks among the threats they were least prepared to address, surpassing cloud threats (40%), third-party breaches (34%), and ransomware (33%).

Conducted across 71 countries and territories, the survey gathered responses from 3,934 business and technology executives and focused on their assessments of preparedness and investment priorities.

Despite increased cybersecurity budgets, only 39% of security, risk, and operations leaders reported having fully formalized and integrated operational continuity plans specifically addressing cyber risks. Data protection also lagged behind, with companies implementing an average of three out of seven data-risk measures examined and only 5% fully implementing all seven.

Inadequate deployment of data classification policies (49%) and data-loss prevention across key channels (48%) leaves organizations vulnerable while pursuing AI investments.

Concerns about autonomous software's role in cyber defense were significant, with only 22% of respondents willing to allow AI agents to execute defensive actions without human approval. AI agents' reliability and technological maturity emerged as the primary barriers to greater autonomy, cited by 55% of respondents. Accountability and explainability of AI decisions were also major concerns, with 44% of chief information security officers identifying workforce skills in AI oversight and governance as a leading obstacle.

Investment priorities in AI revealed a strong emphasis on responsible AI governance (42%), platform hardening (38%), and supply chain security (35%). However, the survey highlights that merely increasing spending on cybersecurity is not sufficient; it must be done smartly, ensuring proper governance, monitoring, and visibility.