AI Adoption Surges as Organizations Lose Visibility and Face Growing Security Risks, AvePoint Report Finds
Jersey City, N.J. – June 29, 2026 – A new report from AvePoint reveals that AI adoption has rapidly scaled into daily enterprise operations, but organizations are struggling with visibility, governance, and security risks as unsanctioned tool usage surges. The State of AI Report: Scaling Trust, Control, and Readiness in the Agentic Era, conducted in partnership with Osterman Research, surveyed 750 enterprise leaders across the Americas, EMEA, and APAC responsible for data security, AI programs, or information management.
Key findings highlight a widening governance gap, with nearly half (46.9%) of global employees using AI agents weekly or daily up from 2025 and work processes incorporating AI agents expected to double within the next year. Organizations anticipate AI agents will replace over 25% of human work in the next 12 months and nearly half within five years. However, cost displacement, not headcount reduction, drives adoption, with ROI measured by reduced manual effort and reallocated human capacity.
Despite rapid deployment, visibility and control lag behind. 21.1% of organizations cannot track unsanctioned AI agent usage, a higher blind spot than the 17.6% lacking visibility into unsanctioned generative AI tools. Top concerns include AI agents making incorrect judgments or bypassing human oversight, while cybersecurity though the most common use case remains high-risk without proper governance.
Confidence in security does not match reality. 82.7% of respondents claim high confidence in preventing unauthorized AI data access, yet 72% of the "very confident" and 62% of the "extremely confident" groups reported AI-related breaches in the past year. Security incidents have risen sharply, with 89.5% of organizations experiencing at least one generative AI-related breach in 2026 (up from 75.1% in 2025) and 88.4% facing AI agent breaches. These gaps have delayed AI deployments for 86.9% of organizations by an average of six months due to data security concerns.
AI-generated data is also reshaping enterprise storage. 35.5% of enterprise data is now AI-generated, expected to reach 42.1% within a year, while 84.1% of organizations manage at least 1 petabyte of data. The combination of aging data (78.1% report at least half is over five years old) and AI-generated content risks compounding governance failures at scale.
To address these challenges, organizations are prioritizing investments in data security for AI training (79.5%) and third-party governance tools, including AI Agent Management Platforms (AMPs). 95.5% have taken action to mitigate AI agent security risks, reducing inaction from 8.3% in 2025 to 2.5% in 2026.
AvePoint’s CEO, Dr. Tianyi Jiang, emphasized that AI’s constraint is no longer model capability but trust visibility, enforceable governance, and control. Without these foundations, rapid deployment increases exposure. The report concludes that scalable AI adoption depends on a "trust layer" operational frameworks ensuring auditable, correctable outcomes.
The full report is available from AvePoint.
AvePoint cybersecurity rating report: https://www.rankiteo.com/company/avepoint
Osterman Research cybersecurity rating report: https://www.rankiteo.com/company/osterman-research-inc.
"id": "AVEOST1782772741",
"linkid": "avepoint, osterman-research-inc.",
"type": "Breach",
"date": "6/2025",
"severity": "85",
"impact": "4",
"explanation": "Attack with significant impact with customers data leaks"{'affected_entities': [{'industry': ['Multiple Industries'],
'location': ['Americas', 'EMEA', 'APAC'],
'name': 'Global Enterprises (Surveyed by AvePoint and '
'Osterman Research)',
'size': 'Enterprises (750 surveyed)',
'type': 'Organizations'}],
'attack_vector': ['Unsanctioned AI Tool Usage',
'Lack of Visibility',
'Insufficient Governance'],
'data_breach': {'type_of_data_compromised': ['AI-generated data',
'Enterprise data']},
'date_publicly_disclosed': '2026-06-29',
'description': 'A report from AvePoint reveals that rapid AI adoption has led '
'to increased security risks, governance gaps, and visibility '
'issues. Nearly half of global employees use AI agents weekly '
'or daily, but 21.1% of organizations cannot track '
'unsanctioned AI agent usage. Security incidents have risen '
'sharply, with 89.5% of organizations experiencing at least '
'one generative AI-related breach in 2026 and 88.4% facing AI '
'agent breaches. Confidence in security does not match '
"reality, as 72% of 'very confident' and 62% of 'extremely "
"confident' organizations reported AI-related breaches in the "
'past year.',
'impact': {'data_compromised': 'AI-generated data and enterprise data',
'operational_impact': 'Delayed AI deployments by an average of six '
'months for 86.9% of organizations',
'systems_affected': ['AI Agents',
'Generative AI Tools',
'Enterprise Data Storage']},
'lessons_learned': "AI adoption requires a 'trust layer' with operational "
'frameworks ensuring auditable, correctable outcomes. '
'Visibility, enforceable governance, and control are '
'critical to mitigating security risks.',
'motivation': ['Cost Displacement', 'Operational Efficiency'],
'post_incident_analysis': {'corrective_actions': ['Implementation of AI Agent '
'Management Platforms '
'(AMPs)',
'Enhanced data security for '
'AI training',
'Operational frameworks for '
'auditable and correctable '
'AI outcomes'],
'root_causes': ['Lack of visibility into '
'unsanctioned AI tool usage',
'Insufficient governance '
'frameworks',
'Overconfidence in security '
'measures']},
'recommendations': ['Invest in data security for AI training',
'Adopt AI Agent Management Platforms (AMPs)',
'Improve visibility into unsanctioned AI tool usage',
'Implement enforceable governance frameworks'],
'references': [{'date_accessed': '2026-06-29',
'source': 'AvePoint State of AI Report: Scaling Trust, '
'Control, and Readiness in the Agentic Era'}],
'response': {'remediation_measures': ['Investments in data security for AI '
'training (79.5%)'],
'third_party_assistance': ['AI Agent Management Platforms (AMPs)',
'Third-Party Governance Tools']},
'title': 'AI-Related Security Breaches Due to Lack of Visibility and '
'Governance',
'type': ['Data Breach', 'Security Misconfiguration', 'Governance Failure'],
'vulnerability_exploited': ['AI Agents Bypassing Human Oversight',
'Unauthorized AI Data Access',
'Lack of AI Agent Management']}