​Just as IT teams have begun to rein in Shadow IT through stronger oversight and access controls, a new — and more complex — challenge has surfaced: Shadow AI.

The explosive growth of AI-powered workplace tools has created both incredible opportunities and unprecedented risks. 

While artificial intelligence is transforming how employees work, the unauthorized use of generative AI (GenAI) applications can expose organizations to serious data security and compliance threats.

What makes this issue more difficult is AI's very nature — self-learning algorithms and deep data integration make it far harder to monitor, trace, and contain. As a result, companies already struggling to manage their information flows are finding it even harder to detect and mitigate AI-driven data exposure.

If the experience with Shadow IT taught us one lesson, it's that governance must evolve as fast as innovation.

The same holds true for AI. 

To stay ahead of emerging threats, today's security leaders must establish proactive AI governance frameworks that address current risks while preparing for the next wave of workplace automation.

The rise of Shadow AI shows that employees aren't the problem — poor visibility is. When companies empower people with secure, governed AI tools, risk turns into opportunity.

— William Reed, Digital Workplace Strategist​

What Is Shadow AI (and Why It's a Growing Risk for Businesses)?

​Shadow AI, short for shadow artificial intelligence, refers to the use of AI tools and platforms without a company's knowledge, approval, or oversight. 

In other words, it's when employees use AI in their daily work — often for speed or convenience — without going through any formal review by IT or security teams.

Recent workplace research shows that over 80% of professionals now use AI tools on the job, and nearly 70% admit to using platforms that haven't been approved by their organization. 

This rise in unregulated AI use is driven by how accessible these tools have become. 

With just a browser and an email address, anyone can start using ChatGPT, Claude, or Gemini to draft emails, summarize reports, or analyze data — no coding skills required.

But here's the problem: these tools often process and store data externally, creating serious risks for privacy, compliance, and intellectual property.

For example, ChatGPT alone now serves over 180 million active users globally, and unless companies configure safeguards, sensitive internal information could be shared beyond intended boundaries.

That's where cloud applications like AgilityPortal comes in.

AgilityPortal helps organizations embrace AI responsibly — giving teams access to AI-powered productivity tools within a secure, centralized environment.

Instead of banning AI and driving it underground, businesses can use AgilityPortal to:

• Enable AI-assisted document search and insights without external exposure.
• Maintain data governance and audit trails for all AI activity.
• Set AI usage policies visible across the digital workplace.
• Educate staff through internal campaigns and policy pages.

By providing a trusted space for AI innovation, AgilityPortal allows companies to harness automation safely — protecting data while empowering teams to work smarter.

The result?

Fewer security risks, better productivity, and an organization that can scale AI adoption without losing control. 

Shadow AI vs. Shadow IT: Understanding the Distinction ​

Although Shadow AI and Shadow IT both involve employees using technology outside official approval channels, the nature of the tools — and the threats they create — are quite different. 

​Aspect	​Shadow AI	​Shadow IT
​What It Is	​The use of artificial intelligence tools or models without organizational oversight or authorization.	​The use of unapproved software, apps, or hardware within company systems.
​Primary Risk Factors	​Exposure of sensitive data, inaccurate AI-generated content, and algorithmic bias leading to poor decisions.	​Weak security controls, unauthorized data access, and potential breaches of company or client data.
​Core Technologies	​Machine learning models, large language models, and generative AI platforms such as ChatGPT, Claude, or Copilot.	​Cloud platforms, SaaS products, file-sharing services, or personal devices used for work.
​Compliance Implications	​Raises unique AI governance and transparency concerns, including accountability, explainability, and ethical data use.	​Involves general IT compliance and data protection laws, such as GDPR or HIPAA.
​Operational Challenges	​Hard to detect and manage due to the hidden nature of AI integrations in everyday apps and workflows.	​Difficult to control when employees install or subscribe to external IT services outside security monitoring.

While Shadow IT has been a known issue for years, Shadow AI is proving far more complex to identify and control. 

Traditional IT tools can be monitored through access logs or device scans, but Shadow AI detection is trickier because AI capabilities are now embedded directly into everyday software — from email to document editors. 

Employees may not even realize they're using AI features that process sensitive data in the background.

Unlike Shadow IT, which is mostly about unauthorized tools, Shadow AI blurs ethical, legal, and operational lines. It's not just a question of access — it's about how data is being generated, shared, and learned from. 

This makes AI governance and oversight one of the most pressing risk management priorities for organizations in 2025 and beyond.

What Causes Shadow AI in the Workplace? 

So, where does shadow AI actually come from?

In most organizations, it's not about employees trying to break the rules — it's about convenience, curiosity, and the lack of clear structure around how AI should be used. 

Shadow AI tends to thrive when three conditions come together:

#1. Easy Access to Generative AI Tools

One big reason shadow AI in the workplace is spreading so fast is how easy these tools are to use. 

You no longer need technical skills to access generative AI platforms like ChatGPT, Gemini, Copilot, or Claude. 

Anyone can generate content, summarize reports, or write code from a browser in seconds — no IT approval required.

That convenience is powerful but risky. Employees can unknowingly introduce shadow AI tools into the organization by using them for daily tasks such as drafting proposals or analyzing feedback. These actions may expose sensitive data, yet they happen outside monitored systems, leaving no audit trail for compliance teams.

The issue grows as AI becomes embedded in tools like Microsoft 365, Slack, and Notion, often operating behind the scenes. 

Even when users don't activate it directly, data may still be processed or analyzed automatically. 

This makes AI risk management harder — and without clear policies and training, businesses risk blurring the line between approved technology and ungoverned AI use.

#2. Weak or Unclear AI Governance

A major driver of shadow AI in the workplace is the lack of a clear AI governance framework. 

Many organizations are still figuring out how to manage this fast-moving technology, leaving gaps in policy and enforcement. Without proper AI usage guidelines, employees are unsure which tools are approved — or what kind of data they can safely share.

This uncertainty opens the door for unauthorized AI tools to slip into everyday workflows, often unnoticed until a security or compliance issue emerges. 

Effective AI governance should include:

• Clear AI usage policies that define approved tools and data types
• Regular reviews and audits of AI activities across departments
• Training to help employees understand safe and responsible AI use
• Oversight from IT, legal, and compliance teams

When organizations skip these steps, they don't just risk data exposure — they lose visibility. 

Without guardrails, AI becomes a black box that can quietly shape decisions, content, and workflows without accountability. 

Strengthening AI policy and governance is essential to balance innovation with control in 2025 and beyond.

#3. Gaps in business efficiency and productivity

Sometimes, teams turn to AI because the approved tools don't fully meet their needs. 

They might use AI to summarize reports, draft emails, or automate repetitive tasks — especially when internal systems feel outdated or slow. This "DIY" approach to innovation often leads to AI compliance and data security risks down the line.

When these three factors overlap — easy access, missing governance, and unmet needs — shadow AI in the workplace flourishes. 

The result is a hidden layer of technology use that can put company data, privacy, and compliance at risk if left unmanaged.

Shadow AI The New Threat (2025 and Beyond)

​Artificial Intelligence has become both the biggest breakthrough and the biggest new risk in the modern workplace.

The promise of instant productivity gains and automation has made generative AI tools the most significant shift since the rise of the internet. 

With how easily they slot into everyday workflows, it's no surprise that employees are embracing them at full speed — often without realizing the potential consequences.

By 2025, AI is built into almost everything. 

Microsoft, Google, and other software vendors are integrating generative AI directly into the tools people use every day — from spreadsheets and email to chat and document creation. That means sensitive company data might be shared with AI systems in the background, even without explicit user intent.

The challenge is that these tools don't just process information — they learn from it. 

Generative AI models analyze massive amounts of data to improve their outputs, and that data often includes financial records, customer information, proprietary code, or internal strategy documents. Once this information leaves an organization's environment, there's little control over how it's stored, shared, or used.

If an AI provider suffers a breach or if cybercriminals manipulate the models, that data could be exposed or repurposed in ways no one intended. 

Even innocent use — like asking an AI to "improve" a client report — could end up contributing private business details to a shared dataset. Over time, this can erode privacy, intellectual property rights, and even competitive advantage.

The reality is that the line between productivity and vulnerability is getting thinner every day. As organizations lean more on AI to get work done, the focus now needs to shift from simply adopting these tools to understanding how they're being used. 

That awareness — knowing where data goes, who interacts with it, and what risks come with it — will define the next phase of responsible AI adoption.

Shadow AI Risks

​Without the right oversight, shadow AI introduces risks that are as wide-reaching as its impact. 

What starts as a small productivity shortcut can quickly become a serious AI security, compliance, and reputation issue. 

Let's look at the top three risks organizations face in 2025 and beyond:

#1. Data Exposure and Loss of Confidentiality ​

​The biggest concern with shadow AI in the workplace is data leakage. Employees often paste snippets of private information — client data, source code, or financial details — into generative AI tools to "speed things up." 

What many don't realize is that some platforms store or reuse this data for model training, which can inadvertently make sensitive information accessible to third parties.

Real-world examples already exist. 

Several large companies, including Samsung and JPMorgan, have had to restrict or ban ChatGPT after employees entered proprietary data into it. 

Even when the data doesn't leave the organization, the lack of encryption or access control on unsanctioned tools increases the risk of data exfiltration and insider leaks.

Key risks include:

• Leakage of confidential or regulated data to third-party AI services
• Loss of intellectual property and trade secrets
• Potential GDPR or privacy law violations from unapproved data sharing

As AI continues to integrate into more SaaS applications, these exposures are expected to rise sharply.

#2. Misinformation and Biased Outputs

Generative AI isn't perfect — it's predictive. 

Models like ChatGPT or Gemini can produce hallucinated or inaccurate responses, especially when uncertain or trained on flawed data. 

When employees rely on these outputs without fact-checking, it can lead to poor decisions, legal errors, and reputational harm.

For instance, in a now-famous case, two lawyers in New York submitted fake legal citations created by ChatGPT, resulting in fines and embarrassment. 

Similarly, biased datasets can lead to biased results — an issue already seen in AI-generated recruitment content or marketing imagery.

Common outcomes include:

• Misinformation entering official reports or decisions
• Reputational damage from AI-generated errors
• Biased or discriminatory content that violates ethics or policy

Unchecked AI bias and misinformation can be just as damaging as a data breach, especially when outputs reach clients or the public.

#3. Non-Compliance and Regulatory Breaches

​AI regulation is evolving fast. 

The EU AI Act, new GDPR AI provisions, and frameworks like NIST's AI Risk Management guidelines are setting stricter expectations for how organizations use and monitor AI. 

Shadow AI — by its very nature — operates outside these controls, making it nearly impossible for businesses to prove compliance if audited.

The result? 

Legal exposure, financial penalties, and reputational harm. Non-compliance with AI-related laws can have ripple effects across industries, especially in finance, healthcare, and government sectors where data protection is tightly regulated.

Key implications:

• Inability to meet data protection and AI governance standards
• Potential fines or legal action from regulatory bodies
• Loss of customer trust due to poor transparency or accountability

As AI adoption accelerates, compliance isn't optional — it's foundational. Organizations that fail to monitor and document AI use risk being blindsided by new rules in 2025 and beyond.

The risks of shadow AI are not hypothetical — they're already happening. 

The solution isn't to block AI entirely but to build visibility, accountability, and education into how it's used. Responsible AI governance frameworks will be the difference between innovation and exposure.

10 Best Practices for Preventing and Managing Shadow AI 

Managing Shadow AI requires more than policies — it demands awareness, collaboration, and a culture that balances innovation with responsibility. 

Here are ten practical strategies organizations can use in 2025 to detect, control, and safely integrate AI into daily operations.

#1. Build a Culture of Responsible AI Adoption​

Shadow AI often thrives because employees see AI as a shortcut, not a threat. 

Promote a culture where responsible AI use is encouraged and openly discussed.

• Host Q&A sessions or "AI office hours" to educate staff.
• Encourage teams to share how they're using AI to identify unapproved tools early.
• Communicate that AI isn't banned — it's guided.
  When people understand why governance matters, compliance becomes second nature.

#2. Define Your Organization's AI Risk Appetite

 Before implementing any AI governance framework, it's essential to understand how much risk your organization is willing to accept. 

Every business has a different comfort level depending on its data sensitivity, regulatory exposure, and industry expectations.

When defining your AI risk appetite, evaluate:

• Regulatory obligations – Understand how laws such as GDPR, HIPAA, and the EU AI Act apply to your organization.
• Data sensitivity – Identify which business units handle confidential data (e.g., finance, HR, customer service) and how AI tools might interact with it.
• Operational dependencies – Assess which business processes rely heavily on AI or automation and how disruption could affect performance.
• Financial and reputational impact – Estimate the potential cost of data leaks, compliance violations, or biased AI outputs.
• Ethical and social considerations – Determine the company's stance on transparency, explainability, and fairness in AI model usage.
• Vendor reliability – Evaluate third-party AI providers for data handling practices, storage policies, and model transparency.
• Incident response readiness – Review whether the organization has protocols in place for AI-related breaches or misuse.

Once your risk appetite is clearly defined, use it to shape your AI risk management strategy. 

Apply strict governance to high-risk use cases (e.g., HR decision-making, financial forecasting) while allowing flexibility for low-risk, productivity-focused tools. This balance enables innovation without compromising trust or compliance.

#3. Strengthen Shadow AI Detection and Visibility

You can't mitigate what you can't see. Invest in Shadow AI detection tools that identify unauthorized AI activity across browsers, APIs, and SaaS apps.

• Track AI traffic patterns, API calls, and plug-ins connected to company systems.
• Maintain a live AI usage inventory across departments.

Visibility is the foundation for effective AI governance — once you know where Shadow AI lives, you can manage it.

#4. Establish a Clear Responsible AI Policy

Create a written policy that defines what's acceptable, restricted, or prohibited when using AI. 

It should include:

• Types of data that can be shared with AI models
• Security and encryption requirements
• Procedures for approving new tools

Review and update this policy regularly to reflect emerging risks and evolving AI laws.

#5. Collaborate Across Departments 

AI governance shouldn't sit with IT alone. 

Bring together legal, HR, security, and operations to align policies and monitoring efforts.

• Shared ownership improves oversight and speeds up decision-making.
• Consistent standards prevent departments from adopting conflicting tools.

Cross-functional collaboration ensures AI usage remains secure and compliant company-wide.

#6. Engage Employees in Governance Design

The best way to eliminate unauthorized AI use is to understand why it happens. 

Conduct surveys or focus groups to learn which tools employees rely on and what problems they're solving.
This approach:

• Reveals gaps in approved technology
• Builds trust between staff and IT
• Makes governance policies more practical and user-centric

#7. Prioritize AI Solutions by Risk and Business Value 

Not every AI tool poses the same level of risk. Classify them by business value and sensitivity.

• Start with low-risk automations (e.g., meeting summaries).
• Apply stronger oversight to high-risk tools handling personal or financial data.
  This prioritization keeps innovation moving while minimizing exposure.

#8. Provide Ongoing Training and Support 

Training is key to sustainable AI compliance. Offer workshops, microlearning modules, or internal newsletters on topics like data sharing, prompt security, and ethical AI use.

Pair this with dedicated support — help desks, quick-reference guides, or internal AI communities — so employees feel confident and informed.

#9. Conduct Regular Shadow AI Audits ​

Schedule routine audits to uncover unauthorized tools, assess risk, and identify trends in AI usage.

• Review AI integrations quarterly or biannually.
• Document findings and update policies accordingly.
  Audits provide a real-world snapshot of how AI is being used and where new risks are emerging.

#10. Continuously Update AI Governance Practices 

AI is evolving faster than most policies can keep up. 

Treat AI governance as a living framework — one that grows with the organization.

• Revisit policies annually or after any major AI tool update.
• Involve multiple stakeholders in reviews to ensure relevance.
• Adaptability is what separates resilient organizations from those caught off guard by new AI risks.

Shadow AI detection and control start with awareness. 

When companies align technology, training, and governance, they transform AI from a hidden risk into a trusted productivity driver. The goal isn't to stop employees from using AI — it's to help them use it safely, transparently, and effectively.

The Future of Shadow AI Governance in 2025 and Beyond

As AI adoption surges across industries, the question for leaders is no longer "Should we use AI?" — but "How do we use it responsibly?"

According to Gartner's 2025 CIO Outlook, nearly 85% of organizations will use some form of generative AI by the end of the year, yet fewer than 40% have formal AI governance or compliance processes in place. 

This widening gap is what fuels the rise of Shadow AI — unmonitored, employee-led use of AI tools that operate beyond IT visibility.

Generative AI has become deeply embedded in modern workflows. From Microsoft 365 Copilot to Google Workspace AI assistants, these tools are now standard in productivity suites. 

However, their convenience also creates blind spots — employees may be sharing confidential data with third-party AI systems without realizing it.

The result? 

Increased exposure to data breaches, model bias, and regulatory non-compliance, all of which fall under the growing field of AI risk management.

• AI Integration Is Becoming Invisible - By late 2025, it's estimated that 60% of workplace tools will include embedded AI features, according to IDC. This makes traditional monitoring nearly impossible — emphasizing the need for continuous Shadow AI detection systems that track usage patterns across cloud, SaaS, and internal networks.
• Regulation Is Catching Up - Governments are tightening oversight. The EU AI Act — the world's first comprehensive AI regulation — will take effect in phases from 2025 to 2026, classifying AI systems by risk. Similar frameworks are emerging in the U.S., Canada, and Singapore. Organizations that fail to implement proactive AI governance frameworks risk financial penalties and reputational damage.
• AI Policy Development Is Becoming Board-Level Priority - A Deloitte 2025 survey found that 72% of executives plan to formalize an AI compliance policy within 12 months, making AI oversight a key pillar of enterprise risk management. Boards now expect transparency in how AI is trained, used, and audited — not just its business benefits.
• The Shift from Restriction to Enablement - Forward-thinking companies are realizing that banning AI doesn't work. Instead, they're focusing on responsible enablement — providing approved, secure AI tools under clear governance rules. This approach helps maintain innovation while minimizing risk, and aligns with emerging AI ethics and data governance standards.
• Data Security and Trust as Competitive Differentiators - In PwC's Global AI Business Survey 2025, 63% of business leaders said that customer trust in AI transparency will directly affect brand loyalty. Companies that can demonstrate strong AI compliance and transparent data use will earn a measurable edge in both market reputation and investor confidence.

Where Shadow AI Governance Is Headed

The next phase of AI maturity will blend automation, compliance, and culture. 

Leading organizations are already investing in:

• Advanced Shadow AI detection tools to identify and categorize AI usage in real time.
• AI governance frameworks that align with ISO/IEC 42001 (AI management systems).
• Risk assessment dashboards integrating AI usage data with cybersecurity metrics.
• Employee awareness programs to make AI governance part of everyday work culture.

In the coming years, Shadow AI management will become as standard as cybersecurity or data protection — an ongoing discipline that safeguards both innovation and integrity.

The future isn't about stopping employees from using AI; it's about building trust, visibility, and accountability around how it's used.

Organizations that act now — combining AI risk management, policy development, and continuous monitoring — will lead the next era of digital transformation safely and confidently.​ 

AgilityPortal: The Secure Way to Harness AI — Without the Risk of Shadow AI 

In a world where AI is everywhere, uncontrolled tools can quickly turn innovation into exposure. 

That's why AgilityPortal was built differently — to help organizations unlock the power of AI securely.

Our AI-powered search engine, built directly into the AgilityPortal workspace, gives your teams instant access to the knowledge they need — from policies to projects — without ever leaving your protected environment. 

Every query stays inside your network, so data never leaves your control.

Unlike public AI tools, AgilityPortal uses governed prompts and controlled access models to ensure that every interaction is compliant, auditable, and aligned with company policies.

• On-platform AI search — Employees get the speed and intelligence of AI without exposing data to external services.
• Secure prompt management — Prompts are logged, encrypted, and never shared beyond your private instance.
• Role-based AI access — Only authorized users can use AI for certain data types or tasks, reducing misuse.
• Shadow AI detection and audit trail — Every AI query and response is tracked for visibility, accountability, and compliance.
• Built-in AI governance — Aligns with global standards like ISO/IEC 42001 and GDPR requirements.

With AgilityPortal, you don't have to choose between innovation and security.

Your teams get the benefits of AI — faster search, smarter insights, and more creative collaboration — all within a zero-leak, policy-controlled environment that prevents Shadow AI from ever taking root.

Because real digital transformation isn't just about using AI — it's about using it responsibly.

See how AI-powered search, governed prompts, and Shadow AI detection work together to keep your organization smart, fast, and fully secure.

• No credit card. 
• No setup headaches. 
• Just 14 days of discovering how seamless and compliant AI can be.

Start Your 14-Day Free Trial and experience the future of responsible AI adoption with AgilityPortal. 

👉 Start Your 14-Day Free Trial

Frequently Asked Questions About Shadow AI Governance and Prevention 

1. Explain Shadow AI in the workplace and its biggest risks for 2025. 

Shadow AI refers to employees using artificial intelligence tools — such as ChatGPT, Gemini, or Copilot — without the company's approval or oversight. 

In 2025, this has become a growing issue as AI becomes embedded into common workplace apps.

The biggest risks of Shadow AI include:

• Data exposure – Sensitive company information shared with external AI platforms.
• Compliance violations – Breaches of GDPR, HIPAA, or the EU AI Act.
• AI bias and misinformation – Inaccurate or unethical outputs influencing decisions.

Organizations can mitigate these risks with structured AI governance frameworks, staff education, and Shadow AI detection systems that monitor AI use safely.

2. Create a checklist for IT and HR leaders to detect and manage Shadow AI use within organizations. 

Here's a quick Shadow AI detection checklist for IT and HR teams:

• Audit browser logs, API calls, and app integrations for unauthorized AI tools.
• Track employee surveys or usage reports to identify popular unapproved AI apps.
• Integrate AI activity monitoring software into network security systems.
• Define a clear AI compliance policy outlining approved platforms and use cases.
• Create a process for employees to request approval for new AI tools.
• Store all AI-generated data securely within the organization's own servers.
  Following this checklist gives leaders visibility, enabling responsible AI adoption while minimizing risks.

3. Write an internal communication plan for introducing responsible AI policies without discouraging innovation. 

When introducing an AI governance policy, tone matters. Focus on empowerment, not restriction.
Here's how:

• Use simple, transparent language — explain why responsible AI matters.
• Host short training sessions that show real productivity benefits.
• Involve early adopters as "AI champions" to encourage peer learning.
• Position governance as an innovation enabler, not a blocker.
• Provide approved, secure alternatives (like AgilityPortal's AI-powered search) so employees don't resort to unapproved tools.

This approach builds trust and promotes a culture of responsible AI use without stifling creativity.

4. Summarize best practices for preventing data leaks and maintaining compliance when using generative AI tools.

To prevent data leaks and ensure AI compliance, organizations should:

• Restrict uploads of sensitive or confidential data to external AI models.
• Use AI-powered search tools that process queries internally.
• Implement consent, encryption, and access control for every AI transaction.
• Conduct quarterly AI governance audits to identify compliance gaps.
• Train employees to recognize unsafe AI prompts and high-risk data sharing.

Combining AI governance frameworks with internal AI alternatives significantly reduces the chance of accidental exposure or misuse.

5. Describe how AI-powered search tools (like AgilityPortal's) can help eliminate Shadow AI by providing secure, governed alternatives. 

Modern platforms like AgilityPortal provide a practical solution to the Shadow AI problem. 

Instead of banning AI, they make it safe to use.

Here's how:

• AI-powered search gives employees fast, accurate answers inside a controlled environment.
• Governed prompts ensure that every AI interaction follows company policy.
• Activity logs and audit trails provide complete visibility for compliance.
• Role-based permissions prevent unauthorized access or misuse.

By offering a secure internal AI experience, companies can replace unapproved Shadow AI tools with transparent, compliant, and efficient systems — keeping innovation alive while maintaining full control.

Resources

These links help readers understand formal governance models, responsible AI policy development, and upcoming regulatory standards.

1. NIST AI Risk Management Framework (AI RMF)
   https://www.nist.gov/itl/ai-risk-management-framework
2. Guide to a Responsible AI Governance Framework – BDO
   https://www.bdo.ca/insights/responsible-ai-guide-a-comprehensive-road-map-to-an-ai-governance-framework
3. AI Governance Library – AIGL
   https://www.aigl.blog/
4. Top 8 AI Governance Platforms for 2025 – Domo
   https://www.domo.com/learn/article/ai-governance-tools

Focused on identifying unauthorized AI use, data exposure prevention, and safe adoption practices.

1. Shadow AI: The Silent Threat to Enterprise Data Security – Security Magazine
   https://www.securitymagazine.com/articles/101382-shadow-ai-the-silent-threat-to-enterprise-data-security
2. Shadow AI Detection with BigID
   https://bigid.com/blog/detecting-shadow-ai-with-bigid/
3. Managing the Security Risks of Unsanctioned AI Tools – UpGuard
   https://www.upguard.com/blog/unsanctioned-ai-tools
4. Wiz Academy: Shadow AI and Its Risks
   https://www.wiz.io/academy/shadow-ai

Authoritative reports and studies that explore the scope, trends, and implications of Shadow AI adoption in enterprises.

1. The Shadow AI Crisis: Why Enterprise Governance Can't Wait – Anaconda
   https://www.anaconda.com/blog/shadow-ai-crisis-in-the-enterprise
2. Shadow AI: Cybersecurity Implications, Opportunities, and Trends – Springer Research
   https://link.springer.com/article/10.1007/s42979-025-03962-x