Call Contact Us

Shadow AI: The Complete Guide to Finding, Governing, and Securing Unsanctioned AI

What it is, where it hides, how to find it, and how to govern it without a ban.

Shadow AI is any AI tool, feature or agent used at work without the organization's formal review, approval or control. It can be a chatbot on a personal account, an AI feature switched on inside software the company already uses, a browser extension, or an AI agent acting with an employee's access. The tools are often useful. The risk is that the organization may not know what data went into them, where it is kept, who can see the output, or what the AI was allowed to do.

What is shadow AI?

Shadow AI is the AI version of an old problem: people adopt a tool because it helps them do the job, before IT, security or compliance has looked at it. In InfoSight's own words, from its piece on agentic AI risk and AI governance: "Shadow AI happens when AI tools, models, agents, or workflows are used without formal review, approval, documentation, or control." It can be entirely well meant. Some familiar examples:

  • A finance manager pastes a draft budget into a free chatbot to tighten the wording.
  • An HR generalist asks an AI tool to soften a performance review, employee name and medical leave included.
  • A sales team signs up for an AI email assistant that reads every inbox it is connected to.
  • An operations lead clicks a new "summarize with AI" button in a long-licensed platform, and nobody has reviewed what it sends where.

Shadow AI is a branch of shadow IT, the older problem of technology used without IT's knowledge. What AI adds is in the data and the output. Information can leave inside the prompt and may be kept or used to improve the vendor's models, depending on the product and its settings, and the answer flows back into real work. AI can also arrive inside tools IT already approved, and AI agents connected to company systems can take actions, sometimes choosing their own steps. Our explainer on what is shadow AI compares the two side by side.

How common is shadow AI?

A survey of office professionals at large companies and one of cybersecurity leaders both find it common. Neither is a census of mid-size organizations, so treat the figures as a signal.

Two-thirds

In a 2026 PagerDuty survey of 1,250 office professionals at companies with at least $500 million in annual revenue, two-thirds of those who had used AI for work said they had used AI tools at work even though they believed company policy did not allow it. In the same survey, 34% said they had entered customer data or information into public AI tools such as ChatGPT, Claude or Gemini.

69%

In a Gartner survey of 302 cybersecurity leaders conducted from March to May 2025, 69% said their organizations suspect or have evidence that employees are using prohibited public generative AI. Gartner predicts that by 2030 more than 40% of enterprises will experience security or compliance incidents linked to unauthorized shadow AI.

The number that matters is the one your own logs and people give you.

Where shadow AI hides

Shadow AI is more than a chatbot website, so blocking one is not enough.

Public chatbots on personal accounts

Free and personal accounts of ChatGPT, Gemini, Claude and similar tools, on a laptop or a phone. OpenAI and Anthropic each say that, by default, they do not use inputs or outputs from their business products to train their models. Microsoft says prompts and responses in Copilot, as used by organizations, are not used to train foundation models. Google says Workspace customer data is not used to train models without the customer's permission. OpenAI, Google and Anthropic each say their consumer apps can use conversations to improve their models, depending on the user's settings. A personal plan can be a consumer plan even when it is paid (OpenAI's Plus and Pro and Anthropic's Pro and Max are examples).

AI inside software you already approved

AI features can be added to email, document, CRM and help desk platforms through routine updates, after the contract was reviewed. Take Microsoft Copilot: Microsoft says it respects your identity model and permissions, so it can surface whatever a user already has permission to open, and loose old sharing settings suddenly matter.

Browser extensions

Depending on the permissions they are granted, writing helpers, summarizers and "chat with this page" add-ons can read what is on screen, including the web apps a user is signed in to. InfoSight's blog has covered a Chrome extension caught harvesting AI chats.

Meeting note-takers

AI note-takers join calls, record them and send out transcripts, sometimes because one attendee linked a calendar. Depending on the tool and its settings, everyone on the call may be recorded and the transcript stored with the vendor. Check recording notice and consent, and where transcripts are kept, before approving one. More on the cyber risks of AI note-takers.

Coding assistants

An unapproved coding assistant can send proprietary source code and configuration files to an outside service, along with any passwords or API keys left inside them. Check what the assistant sends to its provider, and on what contract terms, before approving one.

AI agents and automations

Tools that act, not just answer. An agent connected to email, files or a CRM acts through whatever permissions and credentials it was given, sometimes an app permission or API key nobody inventoried. IBM's 2026 breach research notes that AI agents create new non-human identities that often operate with elevated privileges and can be hard to inventory, monitor and govern. It also found that fewer than half of organizations are actively securing non-human identities. InfoSight's advice is to treat AI agents like identities, not employees.

Shadow AI risks: what can go wrong

Not every unapproved use of AI is dangerous: a social post drafted from public information puts no confidential data at risk, though the output still needs a human read. Risk rises with the sensitivity of the data and the weight of the decision.

tick image

Data leaving the company

Customer records, patient details, financials, contracts and source code go into a tool the company has no agreement with. Depending on the product and its settings, the vendor may keep it or use it to improve its models, and the organization may have little control over it once it is submitted.

tick image

Privacy and regulatory exposure

Rules that protect health, financial and personal information can keep applying when that information goes into an AI tool: the rule reaches the tool because the data went there.

tick image

Intellectual property

Designs, formulas, pricing and unreleased plans pasted into an unapproved outside tool leave the company's control. Under the federal trade secret definition, information is a trade secret only if its owner has taken reasonable measures to keep it secret and the information derives economic value from not being generally known or readily ascertainable by others.

tick image

Wrong answers in real decisions

AI can produce confident, plausible and wrong output: a misread contract clause, an invented citation, a miscalculated rate. When AI involvement is undisclosed, it is hard to enforce a human review of the output.

tick image

New attack paths

The OWASP Top 10 for LLM Applications (2026 edition; LLMs are the large language models behind chatbots) lists prompt injection first: input to a model, including content hidden in a web page, a file or a tool's output, that alters the model's behavior in ways the application's developer did not intend. The same list ranks excessive agency third: damaging actions performed in response to unexpected, ambiguous or manipulated output from a model, usually because the system was given more functionality, permissions or autonomy than the task needed. For shadow AI, that means an unreviewed assistant that reads someone's email or web pages can be steered by instructions hidden in them.

tick image

Costlier breaches

IBM's 2025 Cost of a Data Breach Report, the first edition to study shadow AI, found that one in five organizations in its study reported a breach due to shadow AI, and that organizations with high levels of shadow AI saw breach costs about $670,000 higher than those with little or none.

A well-known case

In May 2023, Bloomberg News reported that Samsung Electronics was banning employee use of generative AI tools such as ChatGPT after discovering that staff had uploaded sensitive code to the platform, according to an internal memo Bloomberg reviewed. According to that memo, Samsung was concerned that data sent to AI platforms is stored on external servers, where it is difficult to retrieve and delete and could end up being disclosed to other users.

How to detect shadow AI in your organization

Start with records you already hold. They can give a first picture before any new tool is bought, though gaps will remain.

Where to lookWhat it showsWho usually holds it
Network, DNS and web proxy logsWhich AI services are being reached, from which devices, and how oftenNetwork or security team
Identity provider app consentsAI tools people signed in to with a work account or allowed to read mail, files and calendars (Microsoft Entra ID, Google Workspace)Identity or Microsoft 365 and Google Workspace admins
SaaS admin consolesWhich AI features are switched on inside the platforms you already license, and for whomThe owner of each platform
Endpoint and browser inventoryInstalled AI apps, browser extensions and coding assistantsEndpoint security admins
Expense and card recordsPaid AI subscriptions bought on corporate cards or claimed as expensesFinance or accounts payable
Cloud and code accountsAPI keys for AI services, and AI tools connected to code repositoriesCloud and development leads

Ask people, with a no-blame survey

Logs may not capture personal phones and home networks, and may not show what data went in. A short survey helps close the gap: ask which AI tools people use, for which tasks and with what kind of information, under a no-blame commitment that HR and legal have agreed in advance. The answers become the first draft of your AI inventory and the list of tools to review. Before anyone reads the contents of employee prompts, agree with legal, HR and privacy on notice, purpose and who may look.

The article on how to detect shadow AI on your network takes each of these records in turn: what it can show, what it misses and where to start.

Ban or govern? Why a ban alone falls short

Blocking every AI website feels decisive, but a ban removes the approved route, not the need. The surveys above show the limit of a rule on its own: in Gartner's, 69% of cybersecurity leaders said their organizations suspect or have evidence that employees are using prohibited public generative AI, and in PagerDuty's, people used AI tools they believed policy did not allow. A ban can also push use toward personal phones and accounts, where the organization has less visibility and no business contract terms of its own, and it can lose sight of the use.

Governing AI makes the opposite trade: a safe, approved route for what people already do, and firm lines only where the data or the decision calls for them. Gartner's advice to CIOs points the same way: set clear enterprise-wide policies for AI tool use, audit regularly for shadow AI activity, and build generative AI risk checks into how they assess SaaS tools. In practice:

  • Approve a small set of tools on business terms, with sign-in through your identity provider.
  • Block the specific services and features that fail review, and say why.
  • Keep the hard line for regulated and confidential data, not for AI as a category.
  • Make requesting a new tool quick, so the approved route is also the easy one.

AI governance basics

Artificial intelligence governance, usually shortened to AI governance, is the set of decisions, rules and checks that settle which AI the organization uses, for what, with which data, and who answers for it. IBM's 2025 Cost of a Data Breach Report found that 63% of breached organizations either had no AI governance policy or were still developing one.

Each building block below needs a named owner, not a new department.

An AI inventory and risk tiers

Keep a living list of every AI tool, feature and agent in use: what it does, who owns it, what data it touches, which accounts or keys it runs under, and whether it is approved. The NIST AI RMF lists this among its governance outcomes: "Mechanisms are in place to inventory AI systems." Then give each use a tier, so the rules scale with the risk.

TierTypical useThe rule
LowDrafting and brainstorming with public or non-sensitive informationAny approved tool, no extra review
ModerateInternal documents and business data that is neither regulated nor confidentialApproved business accounts only, with the data terms reviewed
HighPatient, customer financial, personnel or engineering data, trade secrets and other confidential material, or output that feeds decisions about peopleA named owner, written review, contract terms, logging and human review of output
Not allowedUses the policy rules out, such as regulated data in a personal accountBlocked, with a route to request an exception

What an AI acceptable use policy should contain

Training, monitoring and review

Teach the policy with examples from each team's work, and give it a place alongside the security awareness training people already take. Keep the detection sources running, review the inventory on a schedule, and revisit the policy when a major tool or rule changes.

AI governance frameworks: NIST AI RMF and ISO/IEC 42001 in plain English

Neither framework below is a law. Both are voluntary, and a mid-size organization can borrow from either without a compliance department.

NIST AI Risk Management Framework

The NIST AI Risk Management Framework (AI RMF 1.0), released in January 2023, is a voluntary framework from the U.S. National Institute of Standards and Technology for managing the risks of designing, developing, deploying and using AI systems. It is organized around four functions: Govern, which builds a culture of risk management; Map, which sets the context for an AI system's risks; Measure, which analyzes, assesses and monitors those risks; and Manage, which puts resources against the risks that were mapped and measured.

NIST says the framework is being revised under the White House AI Action Plan. In July 2024 NIST published NIST AI 600-1, a Generative AI Profile of the AI RMF that helps organizations identify the risks specific to generative AI and proposes actions to manage them.

How a mid-size team uses it: to organize the work. Govern is the owner, the policy, the tiers and the inventory; Map is the context of each use: its purpose, its data and who it affects; Measure is detection and testing; Manage is acting on findings.

ISO/IEC 42001

ISO/IEC 42001:2023, published in December 2023, is the international standard for an artificial intelligence management system: it specifies requirements for establishing, implementing, maintaining and continually improving how an organization governs the AI it develops, provides or uses.

ISO says it is for organizations of any size, in any industry, including public sector agencies. Certification to ISO/IEC 42001 is voluntary and is carried out by independent certification bodies, not by ISO itself.

A practical split: use the NIST framework to organize the work, and look at ISO/IEC 42001 when customers or partners ask for a certificate. Organizations with EU customers or staff should also check how the EU AI Act applies to them: a US company can fall under it if it places AI systems on the EU market, or if the output of an AI system it uses is used in the EU. In the US, a few states and cities now regulate specific uses of AI, mostly where it helps make decisions about people, such as hiring, lending, housing or insurance. Texas's law took effect January 1, 2026, and Illinois added AI rules to its Human Rights Act the same day. Neither is about staff using chatbots for everyday work.

Your first 90 days of AI governance

This plan is sized for a small IT team working alongside the day job; a larger or more regulated organization will need longer. The aim by day 90: know what AI is in use, have an approved route for the common cases, and have a policy people have read.

01

Days 1 to 30: find out what is in use

  • Name an executive owner and a small working group: IT, security, legal or compliance, HR and a business lead.
  • Pull the detection sources listed above.
  • Run the no-blame survey.
  • Stop the most serious exposure first, such as regulated data going to personal accounts.
02

Days 31 to 60: decide and approve

  • Build the inventory and give each use a risk tier.
  • Approve a small set of tools on business terms, with company sign-in.
  • Set the AI features in your largest platforms on or off on purpose.
  • Draft the AI acceptable use policy and have legal or compliance review it.
03

Days 61 to 90: roll out and keep watch

  • Publish the policy and approved-tools list, with a simple request route.
  • Train people by role, starting with the teams that hold the most sensitive data.
  • Keep detection running and review the inventory quarterly.
  • Report to leadership: what you found, what changed and what is left.

Shadow AI by industry

The basics are the same everywhere; the data and the rules change.

Hospitals and health systems

The worry is patient information in a prompt, such as a discharge summary pasted in for a plain-language rewrite. Under the HIPAA Privacy Rule, a hospital may let an outside vendor create, receive, maintain or transmit protected health information on its behalf only after it obtains satisfactory assurance, documented in a written contract (a business associate agreement), that the vendor will safeguard the information.

So when a staff member pastes patient information into an AI tool the organization has no business associate agreement with, the rule reaches the tool because of the data, not because the tool is AI. HHS's own guidance on business associates lists a third-party vendor's chatbot on a provider's patient portal, one that handles patients' health information, as an example of a business associate.

Banks and credit unions

The federal banking agencies' Interagency Guidelines Establishing Information Security Standards (the FDIC and the Federal Reserve publish the same guidelines for the banks they supervise) tell each bank to exercise due diligence in selecting its service providers, require them by contract to protect customer information, and monitor them where its risk assessment indicates. NCUA's guidelines direct the same oversight for federally insured credit unions, and NCUA says it has no AI-specific rules but that its existing regulations are technology-neutral and apply to AI. An AI tool that receives customer or member information is a service provider question.

Non-bank lenders and credit unions without federal deposit insurance, where the FTC is their regulator, answer to the FTC's Safeguards Rule, which also requires them to oversee their service providers.

Manufacturers

The exposure is designs, process recipes, supplier pricing and engineering data. Because trade secret protection depends in part on reasonable measures to keep information secret, make the rule plain: engineering data goes only into tools approved for it.

Plants add a further concern. AI connected to plant systems needs the same caution as any other change near operations; InfoSight's blog explains why AI in OT is a trust and safety problem first.

State and local government

Many states now set rules for how their own agencies and employees may use generative AI. Washington, for example, bars agencies from entering non-public data into unlicensed AI tools without a contract.

Public records are the other concern: North Carolina's guidance warns that information entered into a publicly available AI tool is considered released to the public and may be subject to public records requests. City and county IT teams can borrow from both.

What an AI exposure assessment looks at

An external AI exposure assessment applies the discovery and governance work described above, using an outside team. InfoSight's AI Governance & Exposure Assessment is built for "Any organization handling sensitive data in a regulated environment." InfoSight lists "Where is AI actually being used in our organization?" among the questions it hears in nearly every first conversation. Its published scope:

What you get, as the page lists it: an Executive Summary, a Technical Findings Report, a Mitigator® Risk Dashboard, an AI Governance Roadmap, Quick Win Recommendations and a Readout Meeting. For AI your organization deploys or builds, AI penetration testing tests how an AI system behaves under adversarial pressure: whether it can be persuaded to act against its intended purpose. InfoSight also runs employee security awareness training, which covers phishing, social engineering and AI-driven deepfakes.

Shadow AI questions people ask

It depends on the account and the approval. ChatGPT used through a business account your organization approved and manages, for uses its policy allows, is sanctioned AI. A personal account used for work without the organization's approval, or company data pasted into an AI tool the organization has not formally reviewed or approved, is shadow AI.

Start with records you already hold: network, DNS and proxy logs, app consents in your identity provider, SaaS admin settings, endpoint and browser-extension inventories, and expense records. Then run a short, no-blame survey to catch what logs miss, such as AI used on personal phones.

A blanket ban removes the approved route but not the need, so use may move to personal phones and accounts you cannot see. Approve a small set of tools on business terms, block the services that fail review, and keep firm lines for regulated and confidential data.

An AI acceptable use policy should cover its scope, the approved tools and how to request new ones, the data that may never go into an AI tool, account rules, when a person must review AI output, who may connect AI tools or agents to company systems, disclosure and reporting, and who owns the policy.

No single law makes shadow AI illegal. Whether a particular use breaks a rule depends on the data involved, what the AI is used to decide, and the laws, contracts and policies that apply. When regulated information such as patient records or customer financial information goes into an AI tool, the rules that protect that information still apply. A few states and cities also regulate AI used in decisions about people, such as hiring. This is general information, not legal advice.

AI governance needs one executive who answers for it, such as the CIO, the CISO or the head of risk or compliance, with a working group from IT, security, legal, HR and the business. Each approved tool also needs a named owner.

AI governance is the set of decisions, rules and checks that settle which AI an organization uses, for what, with which data, and who answers for it. The NIST AI Risk Management Framework and ISO/IEC 42001 give that work a structure.

Start with a 30-minute AI Risk Conversation

No audit language. No commitment.

Prefer the phone? Call 305.828.1003

Ready to Slash Your Cyber Risk in 30 Days?

Schedule a consultation from one of our expert advisors.

No spam—one security expert follow‑up, that’s it.

Want to Receive our Newsletter?

Stay informed of the latest cyber trends.