Free Assessment

Table of Contents

Reach SOC 2 Compliance in 6 Weeks or Less.

Request a Demo

Home

  /

  / EU AI Act Article 50: AI Transparency Rules from August 2026

EU AI Act Article 50: AI Transparency Rules from August 2026

Picture of Pedro Dias
Copy Link

The EU AI Act’s transparency requirements take effect on 2 August 2026, and most of the companies they cover still think the rules are not their problem. Article 50 applies to any business that publishes AI-generated content or runs an AI system that talks to people in the EU. That includes the marketing team generating campaign images and the support team running a chatbot. It also covers the AI agents you’ve wired into customer email. Penalties reach €15 million or 3% of total worldwide annual turnover, whichever is higher, and you don’t need an office in Europe to be in scope. If your content or your chatbot reaches EU users, the obligations reach you.

In a nutshell: if you publish AI-generated images or video, deploy chatbots or AI agents that interact with EU users, or publish AI-written text on matters of public interest, then yes, the EU AI Act applies, starting 2 August 2026.

A quick word on the “AI Act delay” headlines. The Digital Omnibus package did push the high-risk system deadlines back, in some cases by more than a year, but it did not move the deployer obligations in Article 50. Companies that read those headlines and stood down their AI Act work made an expensive mistake, because the rules most likely to touch an ordinary business are the ones that stayed on the calendar.

What Article 50 Actually Requires

Article 50 of the AI Act sets out transparency obligations in four situations. In plain English:

  • Tell people when they’re talking to AI. Systems designed to interact directly with people — chatbots, voice assistants, and AI agents — must make clear that the user is dealing with AI, unless that’s already obvious.
  • Mark AI-generated content so machines can detect it. Providers of generative AI systems must mark outputs in a machine-readable format, typically through metadata and watermarking, so the content is detectable as artificially generated.
  • Label deepfakes. Anyone deploying AI to generate or manipulate image, audio, or video content that resembles real people, places, objects, or events, and could falsely appear authentic, must disclose that the content is artificial.
  • Label AI-generated text on matters of public interest. Text published to inform the public must carry a label if AI-generated or manipulated, unless a human reviewed it and a person or organization holds editorial responsibility for it.

Article 50 also covers emotion recognition and biometric categorization systems, which carry their own disclosure duties. Far fewer businesses run into those, so this article sticks to the four above.

The distinction running through all of this is provider vs deployer. The provider builds or supplies the AI system. The deployer uses it professionally. Most companies reading this are deployers.

Let Axipro help you build a business continuity plan that's practical, compliant, and audit-ready.

Schedule Your Free Assessment Today

Schedule a consultation

If You Use AI-Generated Images

Realistic AI images sit closer to the deepfake rules than most marketing teams assume. The Act’s definition covers content depicting people, objects, places, and events that could falsely appear authentic to a viewer, which describes a large share of what image generators produce for campaigns, social posts, and landing pages.

So what does “clearly and distinguishably labeled” mean? The threshold is best described by its failures: a tiny disclosure hidden in the website footer doesn’t qualify. Neither does a faint label on an image, a label that flashes for an instant in a video, or a disclosure buried in your terms and conditions.

The label has to be visible right where someone sees the content, and it has to meet accessibility standards so people with disabilities can perceive it too.

The Code of Practice proposes a standardized “AI” visual label, localized per language (“KI” in German, “IA” in French). It also draws a useful line between fully AI-generated content and AI-assisted content, with lighter requirements for the latter. A designer who used AI to extend a background is in a different position from a team publishing a fully synthetic image of a person who doesn’t exist.

Important: The deepfake duty doesn’t care about intent. A flattering, harmless AI image of your CEO at an event that never happened is still a deepfake under the Act. Marketing teams generate this kind of content casually. From August, every one of those images needs a label.

EU AI Act Article 50

If You Deploy AI Agents or Chatbots

The rule itself is simple: people must know they’re dealing with AI. The provider carries the design obligation, but as the deployer you’re the one putting the system in front of your customers, and you’re the one an EU regulator will contact if your branded assistant pretends to be human.

The Act contains an exception for cases where it’s “obvious” the user is talking to AI, judged from the perspective of a reasonably well-informed and observant person. Don’t lean on it. What’s obvious to your product team isn’t obvious to every customer, and the human-sounding voice agents and email-writing AI agents rolling out right now are designed specifically to not feel like software. If an AI agent negotiates a renewal over email or handles a support ticket end to end, disclose it.

Pro Tip: Put the Disclosure at the Start of the Interaction

Put the disclosure at the start of the interaction, in the interface itself: "You're chatting with an AI assistant." A line in your privacy policy doesn't meet the standard, and a disclosure that appears after the conversation ends is worthless. For voice agents, say it up front in the greeting.

What Your AI Vendors Owe You

The machine-readable marking obligation in Article 50(2) sits with providers — the companies supplying your generative AI tools. The final Code of Practice expects providers to apply at least two layers of marking where necessary, such as embedded metadata combined with watermarking, and to offer detection mechanisms so deployers, authorities, and researchers can verify whether a piece of content came from AI.

One timing caveat: the Digital Omnibus gives generative AI systems already on the market before 2 August 2026 until 2 December 2026 to comply with the marking requirement. Every other Article 50 obligation stays on the August date.

The practical consequence lands in your contracts. The hard part of Article 50 is deciding who is the provider and who is the deployer at each link in the supply chain, and your vendor agreements need to say which party holds each duty. Ask your generative AI vendors three things: what marking technology they apply, how you can access their detection tooling, and whether they’ve signed the Code of Practice.

Insider Note: The four-month marking grace period creates an awkward window. Between August and December 2026, some of the tools you use daily won’t watermark their output yet, while your deployer duties apply in full from day one. So don’t build your labeling process on the assumption that vendor watermarks have you covered. Label at your own layer, visibly, and treat vendor-side marking as reinforcement rather than the plan.

The Code of Practice: Voluntary, but the De Facto Standard

The European Commission published the final Code of Practice on marking and labelling of AI-generated content, drafted by independent experts with input from more than 180 stakeholders. It has two sections: one for providers, covering marking and detection, and one for deployers, covering the labeling of deepfakes and public-interest text.

Signing is voluntary. Ignoring it is unwise. The Code is the only official, concrete answer to “what does compliance actually look like,” and regulators will most likely use it as their reference point when they judge whether a company made a good-faith effort. The Commission has also prepared guidelines interpreting Article 50’s scope, definitions, and exceptions, with the final version expected around the application date.

Behind on This? A 7-Day Catch-Up Plan

If today is the first time anyone in your company has looked hard at Article 50, a week of focused work covers most of the exposure.

Days 1 and 2 — Inventory your AI content flows.
List every place AI-generated text, images, audio, or video reaches an EU audience, plus every chatbot, voice assistant, or AI agent that interacts with EU users. Count what marketing, sales, and support produce with AI tools, alongside anything built into your product.

Day 3 — Assign roles.
For each tool and use case, decide whether you’re the provider or the deployer, and note which Article 50 duty attaches. This one exercise clears up most of the confusion.

Day 4 — Fix your visual content.
Add clear labels to realistic AI-generated images and video, and audible disclosures to synthetic audio, following the Code’s guidance on visible, persistent labels.

Day 5 — Fix your interactions.
Add disclosures to every chatbot, assistant, and agent at the start of the interaction, and check they meet accessibility requirements.

Day 6 — Fix your contracts.
Write to your generative AI vendors, request their marking and detection documentation, and update agreements to state who holds each duty.

Day 7 — Document everything.
Document everything — record what you inventoried, the roles you assigned, the labels you applied, and the reasoning behind each judgment call. If a regulator ever asks, this record is your first line of defense.

Let Axipro help you build a business continuity plan that's practical, compliant, and audit-ready.

Schedule Your Free Assessment Today

Schedule a consultation

Where Article 50 Fits in Your AI Governance Program

Nothing in that checklist should live as a one-off project. An AI system inventory, assigned roles and responsibilities, transparency controls, vendor management, and documented decisions are all core requirements of ISO 42001, the international standard for AI management systems. Companies that already run an ISO 42001 program will find that Article 50 mostly falls out of controls they operate anyway. Companies scrambling this week are usually the ones without a governance structure to hang the work on.

That’s the real lesson of this deadline. The EU has more AI rules coming: the marking obligation lands in December, and the high-risk regime arrives later. Treating each deadline as a fire drill doesn’t scale. Axipro helps companies build ISO 42001-aligned AI governance programs that absorb requirements like Article 50 as routine control updates instead of emergencies. If the checklist above turned up gaps you can’t close on your own, that’s the work we do. Talk to us.

The short version: Article 50’s transparency requirements apply from 2 August 2026 to any business whose AI-generated content or AI systems reach EU users. Deployers must label deepfakes and public-interest AI text and make sure people know when they’re talking to AI. Providers must mark synthetic content, from December for tools already on the market. The final Code of Practice explains what good looks like.

Heavy AI use isn’t what puts a company at risk here. The risk comes from never writing down where that AI shows up.

Frequently Asked Questions

Does Article 50 apply to companies outside the EU?

Yes. The obligations follow the content and the users, not your headquarters. A US or GCC company publishing AI-generated campaigns to EU audiences, or running a chatbot that serves EU customers, is in scope.

Partly. The Digital Omnibus moved the high-risk system deadlines back and gave providers of generative AI systems already on the market until 2 December 2026 for the machine-readable marking duty. The deployer obligations — deepfake labeling, public-interest text labeling, and chatbot disclosure — all apply from 2 August 2026 as originally scheduled.

No. The text-labeling duty covers AI-generated or AI-manipulated text published to inform the public on matters of public interest, and it includes a carve-out where a human reviewed the text and someone holds editorial responsibility for it. Ordinary marketing content that goes through human review generally doesn’t need a label. Realistic synthetic images and video face stricter treatment.

No, it’s voluntary. But it’s the Commission-backed description of what compliance looks like in practice, and following it is the most defensible way to show good faith. Most companies should treat it as the standard.

Axipro Author

Picture of Pedro Dias

Pedro Dias

Pedro has been writing online for over 10 years. With experience in all things programming, cyber security, and compliance, he is our editor-in-chief at Axipro.
Copy Link

Blog Highlights

Explore More Articles

Read More Blogs

EU AI Act Recruitment Rules: What to Know for 2027

The EU AI Act names recruitment AI as high-risk. Annex III explicitly lists AI systems used for recruitment, candidate selection, and employment decisions, which pulls CV screeners, video interview platforms, and assessment tools into the most demanding compliance regime the Act contains. The original compliance date for these systems was August 2, 2026. In June 2026, the EU’s Digital Omnibus moved the deadline to December 2, 2027, a 16-month extension that has led many HR and talent teams to shelve the topic entirely. That’s a mistake, for two reasons. First, one rule that directly affects recruitment technology is already in force: the ban on emotion recognition in the workplace has applied since February 2, 2025, and it catches features still shipping in some video interview products today. Second, the deferred obligations didn’t shrink. Conformity assessments, human oversight design, bias monitoring, and documentation all still arrive in full, and the practical work of auditing a recruitment stack, renegotiating vendor contracts, and training hiring teams routinely takes a year or more. Here’s what the EU AI Act actually requires of employers and vendors using recruitment tools, on the timeline that now applies. Why Recruitment Tools Are Classified as High-Risk Under the EU AI Act​ Definition of High-Risk AI Systems in Hiring​ The Act takes a list-based approach. Annex III, point 4, designates as high-risk any AI system intended for the recruitment or selection of natural persons, including placing targeted job advertisements, analyzing and filtering applications, and evaluating candidates. The same point covers AI used for decisions on promotion, termination, task allocation, and monitoring of workers, so the classification follows the tool through the entire employment lifecycle, not just the hiring funnel. The reasoning is straightforward: hiring decisions shape access to livelihoods, and algorithmic discrimination in hiring is well documented. The European Commission’s regulatory framework for AI treats employment as one of the areas where an AI error or bias causes serious harm to fundamental rights. That’s the test for the high-risk tier. Types of Recruitment Tools Affected In practice, the high-risk classification captures most of the modern recruitment stack: CV and resume screeners that rank or filter applicants, video interview platforms that score responses or delivery, psychometric and skills assessment tools that produce scores feeding a hiring decision, sourcing and matching algorithms that decide which candidates a recruiter sees, and programmatic job ad targeting systems that determine who sees a vacancy at all. If the system’s output materially influences who advances and who does not, assume high-risk until proven otherwise. Important: Emotion recognition is not high-risk in the workplace. It is prohibited. Article 5 bans AI systems that infer emotions of people in the workplace (outside narrow medical and safety cases), and that ban has applied since February 2025 with the Act’s top penalty tier attached. If your video interview vendor markets “engagement scoring” or “sentiment analysis” of candidates, that feature needs to be switched off for EU hiring now, not in 2027. Recruitment Tools That May Fall Outside High-Risk Classification Not everything in the HR stack qualifies. The Act carves out systems performing narrow procedural tasks that do not materially influence decision outcomes. An applicant tracking system that stores applications, schedules interviews, and sends templated emails is a database with a workflow, not a high-risk AI system. The same goes for tools that transcribe interviews without scoring them, deduplicate candidate records, or generate first drafts of job descriptions for a human to edit. The line is decision influence: the moment a tool ranks, scores, filters, or recommends candidates, it crosses into Annex III territory. Deployers who rely on an exemption must be able to document that assessment, so “we decided it doesn’t count” needs to exist on paper. Extraterritorial Scope: Which Employers Are Covered The Act applies to providers placing AI systems on the EU market and to deployers established in the EU, but it also reaches further: it covers providers and deployers located outside the EU where the output of the system is used in the EU. For recruitment, the consequence is blunt. A US or UK company with no EU entity that uses an AI screener to filter applicants for roles based in Berlin or Dublin, or that screens candidates located in the EU, is using the system’s output in the Union. Brexit doesn’t move UK employers out of scope when they hire into or from the EU. Providers vs. Deployers of Recruitment AI Tools The Act splits obligations between the provider (the vendor that develops the tool and places it on the market) and the deployer (the employer using it). Most employers are deployers, and deployer obligations are lighter but real. One common trap: an employer that substantially modifies a high-risk system, or puts its own name on it, can be reclassified as a provider and inherit the full provider stack. Heavy customization of a screening model, or fine-tuning it on your own hiring data, can be enough to trigger this. Key Obligations for Employers Using AI Recruitment Tools Human Oversight in Automated Hiring Decisions Deployers must assign oversight of the system to people with the competence, training, and authority to intervene. That last word matters. A recruiter who rubber-stamps whatever the ranking algorithm produces, because nobody has time to review 800 rejected CVs, doesn’t count as oversight. Regulators and courts will look at whether the human could genuinely override the system and whether they ever did. Designing review checkpoints where a person can meaningfully change the outcome, and logging when they do, is the core of compliant deployment. Transparency Requirements Toward Candidates Employers must inform workers and their representatives before putting a high-risk AI system into use at work, and candidates subjected to such a system must be told it is being used. In countries with works councils, such as Germany, this obligation lands on top of existing co-determination rights, so employee representatives may need to be consulted before the tool goes live rather than just told afterward. Burying an AI disclosure in a privacy policy paragraph is unlikely to survive scrutiny.

Read more

4 SOC 2 Controls Auditors Reject Even When Your Compliance Tool Says Passing

A green dashboard is not an audit opinion. Compliance automation platforms like Vanta, Drata, Secureframe, and Hyperproof have made SOC 2 readiness faster and cheaper, but every audit cycle produces the same pattern: controls that sat at “passing” for months come back from the auditor with exceptions or requests for re-testing. The four controls below account for a disproportionate share of those rejections, and they all fail for the same underlying reason. The tool confirmed that evidence exists. The auditor tested whether the control actually operated. This article walks through each of the four: what auditors reject, why, and how to fix the evidence before fieldwork starts. Why Compliance Tools Show “Passing” But Auditors Still Reject Controls​ The Gap Between Automated Checks and Auditor Judgment Compliance platforms run continuous control monitoring: API calls that check whether a configuration exists, a document is uploaded, or a task is marked done. That’s real value. It catches drift, keeps evidence in one place, and saves weeks of screenshot collection. An audit is a different exercise. A SOC 2 examination is an attestation performed by a CPA firm under AICPA standards, and the auditor’s job is to form an independent opinion on whether your controls met the Trust Services Criteria. That opinion rests on professional judgment, not on whether an API integration returned a 200 response. What “Passing” Actually Means in Your Compliance Dashboard​ When a control shows “passing,” the platform is telling you one narrow thing: at the moment of the last scan, an automated test found the artifact or setting it was programmed to look for: MFA enforced in the identity provider, a policy document uploaded, a training campaign sitting at 100%. The test says nothing about whether the underlying process ran the way your control narrative claims it did, or whether it ran that way across the whole audit period. How Auditors Evaluate Controls Beyond the Checkbox Auditors test two dimensions. Design effectiveness asks whether the control, as described, would meet the criterion if it worked as intended. Operating effectiveness, the core of a SOC 2 Type 2 report, asks whether it actually did throughout the audit period. To answer that, the auditor pulls a population (every access review, every change, every new hire in the period), selects a sample, and inspects the evidence item by item. A dashboard status feeds into that process. It doesn’t replace it. Insider Note: Auditors increasingly ask for evidence outside the compliance platform precisely because they know what the platform auto-collects. If every artifact you produce comes from the same tool export, expect the auditor to independently pull the population from the source system and compare. Discrepancies between the two are one of the fastest routes to an exception. Control #1: Access Reviews That Automation Marks Complete but Auditors Reject Why Auditors Reject Automated Access Review Evidence​ User access reviews sit under the logical access criteria (CC6.1 through CC6.3), and they are the single most common source of audit exceptions we see. The typical failure: the platform generated a user list, someone clicked “complete,” and the dashboard turned green. The auditor then asks a simple question the evidence can’t answer: what did the reviewer actually decide? The Missing Element: Documented Reviewer Judgment​ An access review is a judgment control. Someone with knowledge of the system must look at each account and confirm the access is still appropriate for the person’s role. A timestamped task closure proves the task was closed. It doesn’t prove anyone assessed anything, and an “approve all” review completed in ninety seconds gets exactly the skepticism it deserves. What Auditors Actually Want to See in Access Review Evidence Auditors look for four things: The full population of accounts at the time of review (including service accounts and admin roles), Evidence of who reviewed it and when, explicit dispositions per account or group (retain, modify, revoke), and Proof that flagged access was actually removed. That last item, the deprovisioning ticket showing revocation within a defined window, is the piece most companies can’t produce. How to Fix Your Access Review Control Before the Audit​ Assign a named control owner per in-scope system, run reviews quarterly, and require reviewers to record a disposition for every line, not a blanket approval. When access is revoked, link the removal ticket to the review record. If a quarter was missed, don’t backfill it. Document it honestly and show the remediation, because auditors treat fabricated retroactive evidence far more severely than a disclosed gap. Control #2: Change Management Approvals That Pass Automated Scans​ Why Ticket Closure Isn’t Proof of Approval​ Change management (CC8.1) automation typically verifies that production changes link to a ticket and the ticket is closed. Auditors test something stricter: that each sampled change was approved by an authorized person before deployment. An approval added after the merge, or a ticket closed by the same engineer who wrote the code, fails that test even though every automated check came back green. The Segregation of Duties Problem Automation Misses Segregation of duties is the requirement that no single person can develop, approve, and deploy the same change. NIST’s SP 800-53 control catalog treats it as a foundational access control principle, and SOC 2 auditors apply the same logic. Small engineering teams trip on this constantly. Self-approved pull requests, admins who can bypass branch protection, direct pushes to main: a scanner sees “changes with tickets” while an auditor sees SoD violations. Emergency Changes and Retroactive Approvals: Common Rejection Triggers​ Every audit period contains hotfixes. Auditors don’t reject emergency changes. They reject emergency changes with no documented post-hoc review. If your policy says urgent changes get retroactive approval within two business days, the auditor will sample your emergency changes and check exactly that. No policy, or a policy nobody followed, produces an exception. Rebuilding Change Management Evidence Auditors Will Accept​ Enforce the control technically: branch protection requiring at least one independent reviewer, no admin bypass, and deploy pipelines that only run from protected branches. Then write the emergency change procedure down and generate the review artifact every time it fires.

Read more
AI Tool Usage Tracking

AI Tool Usage Tracking: Why CASB & DLP Miss Shadow AI

One in five breached organizations last year traced the incident to shadow AI, and those breaches cost an average of $670,000 more than standard incidents, according to IBM’s 2025 Cost of a Data Breach Report. The worst part is that most of those organizations already ran a CASB, a DLP program, or both. The tools were on, but the traffic still got through. That’s the visibility gap this article is about. AI tool usage tracking isn’t the same problem as SaaS discovery, and the security stack built for the SaaS era misses most of what matters about AI. Below, we break down what tracking actually requires, where CASB and DLP fail, which categories of AI usage slip through, and what a stack that works looks like in 2026. What AI Tool Usage Tracking Actually Means Most teams that say they “track AI usage” mean they can see that someone visited chat.openai.com. That is app discovery, and it answers almost none of the questions a security or governance team actually needs answered. Beyond App Discovery: Tracking Prompts, Data Flows, and Model Interactions Real tracking covers three layers. First, which tools are in use: chatbots, copilots, coding assistants, embedded SaaS features, agents. Second, what data moves: the content of prompts, uploaded files, and pasted context, mapped against data classifications. Third, how models behave in your environment: which endpoints get called, which OAuth grants exist, which agents hold standing permissions. Seeing that an employee opened ChatGPT gets you nowhere. What you actually need to know is whether they pasted a customer contract into a personal account while they were there. The Difference Between Detection, Monitoring, and Continuous Tracking Detection is a point-in-time answer to “what AI is here?” Monitoring watches known tools on an ongoing basis. Continuous tracking is broader: it assumes the inventory changes weekly, correlates identity, data, and endpoint signals over time, and feeds a governance program rather than a one-off report. Frameworks such as the NIST AI Risk Management Framework and ISO 42001 assume the third mode. A discovery scan from last quarter won’t satisfy an auditor, and it certainly won’t slow down an attacker. Why Traditional SaaS Monitoring Falls Short for AI SaaS monitoring was built around a stable premise: an app is a destination with a domain, a login, and an admin console. AI breaks that premise in several ways at once. The risky activity is the content of an interaction, not the visit. The tool often isn’t a destination at all but a feature inside an app you already sanctioned. And increasingly the “user” isn’t a person but an agent acting on delegated credentials. Why CASB Misses Shadow AI Usage The Cloud Access Security Broker sits between users and cloud services to enforce policy, and for classic SaaS governance it still earns its keep. AI has structural blind spots that no amount of tuning can fix. CASBs Were Built for SaaS Apps, Not Model Endpoints A CASB catalog maps domains to applications with risk scores. AI usage doesn’t resolve neatly to a domain. The same api.openai.com endpoint serves a sanctioned enterprise deployment, a developer’s weekend experiment, and a data-leaking browser extension, and the catalog sees one “app”. Meanwhile, new model endpoints, wrappers, and niche AI tools appear faster than any vendor catalog can keep up with. Gartner research from late 2025 found 69% of organizations already suspect or have evidence that employees use prohibited public generative AI tools, catalog or no catalog. Blind Spots in Encrypted API Traffic to LLM Providers Prompt content travels over TLS. Without full TLS inspection, a CASB sees connection metadata: destination, volume, timing. It can’t see that the payload contained source code or patient records. And full TLS inspection is harder than the datasheet implies. Certificate pinning breaks it for many native apps and CLI tools, legal and works-council constraints limit it in the EU, and most organizations carve out broad exemption lists that AI traffic happily rides through. The OAuth and Embedded AI Problem CASBs Can’t See When an employee grants an AI meeting-notes tool access to their calendar and mailbox via OAuth, no proxy is involved at all. The vendor’s servers communicate directly with Microsoft’s or Google’s APIs using a persistent token. The same applies to AI features embedded inside sanctioned SaaS, think Notion AI, Slack AI, or Salesforce Einstein. The CASB sees approved traffic to an approved app, while the AI processing happening inside it, and whichever sub-processor it forwards data to, stays invisible. Personal Accounts and BYO-AI Bypass CASB Proxies Netskope’s 2026 Cloud and Threat Report found that nearly half of employees who use generative AI at work do so through personal accounts. Personal accounts on managed devices are hard enough; personal accounts on personal devices, home networks, and mobile connections never touch the corporate proxy path at all. Tenant restrictions help for a handful of major providers and do nothing for the long tail. Browser-Based and Extension-Delivered AI Escape Network Inspection AI browser extensions read page content and form inputs locally, then exfiltrate via their own backend, often to generic cloud infrastructure that categorizes as “technology” rather than “AI”. From the network’s view, it is routine HTTPS to a CDN. The riskiest interaction, an extension scraping everything an employee views, produces the most boring traffic signature. Insider Note: In AI governance readiness assessments, the OAuth grant review is where clients get the biggest surprise. We routinely find dozens of AI tools holding live mail, calendar, or drive scopes that nobody in IT ever approved, granted by employees who abandoned the tool (and sometimes the company) months earlier. The tokens keep working anyway. Why DLP Fails to Catch Shadow AI Data Exposure DLP has the opposite problem. It can sometimes see content, but it doesn’t understand it, and AI interactions defeat the pattern matching it depends on. Prompt-Based Data Loss Doesn’t Match DLP Signature Patterns DLP fires on signatures: credit card regexes, SSN formats, keyword dictionaries, file fingerprints. Sensitive prompts rarely look like that. “Summarize why we’re losing

Read more