Table of Contents

Reach SOC 2 Compliance in 6 Weeks or Less.

  /

  / CMMC Level 1 Requirements: A Complete Guide

CMMC Level 1 Requirements: A Complete Guide

Phase 1 of the Cybersecurity Maturity Model Certification program went live on November 10, 2025. From that date, the Department of Defense can write CMMC requirements directly into new solicitations, and contractors who handle even basic government data cannot win awards without a current CMMC status in the Supplier Performance Risk System (SPRS). For roughly 63 percent of the Defense Industrial Base, that means Level 1: 15 foundational safeguards, an annual self-assessment, and a signed affirmation from a senior official.

Level 1 is the smallest version of CMMC. It is also the one most contractors are about to encounter first, and the one with the highest false-confidence rate. This guide covers every requirement, every assessment objective, and every step from scoping to SPRS submission.

CMMC Level 1 Requirements Guide

What Is CMMC Level 1?

CMMC Level 1 (Foundational) is the entry tier of the Cybersecurity Maturity Model Certification program, codified in 32 CFR Part 170. It requires defense contractors who handle Federal Contract Information (FCI) to implement 15 basic safeguarding practices and to confirm that implementation through an annual self-assessment.

The 15 practices come directly from FAR 52.204-21, Basic Safeguarding of Covered Contractor Information Systems, a clause that has technically applied to federal contractors since 2016. What CMMC added is an assessment methodology and a verification mechanism. Until CMMC, no one was checking whether contractors actually did the 15 things they were contractually obligated to do. Under the final CMMC Program Rule, effective December 16, 2024, that gap is closed.

Earlier CMMC drafts described Level 1 as a 17-practice framework because three physical-protection requirements were listed separately. The final rule consolidates them, and the official count now sits at 15 practices with 17 underlying assessment objectives drawn from NIST SP 800-171A. Both numbers are correct, depending on which level of granularity you are working at.

What Is the Purpose of CMMC Level 1?

The purpose is narrow and specific: to protect FCI from unauthorized disclosure. 

FCI is information the federal government either generates or receives during contract performance that is not intended for public release. Think proposal correspondence, delivery schedules, performance reports, and routine contract communications. None of it is classified. None of it is even particularly sensitive in the traditional sense. But aggregated across thousands of contractors and exposed to adversaries, it gives a meaningful picture of what the U.S. government is buying, from whom, and on what timeline.

Level 1 exists because too much of the Defense Industrial Base was failing to apply even basic hygiene to that data. CMMC Level 1 turns inconsistent expectations into a yearly verification cycle.

CMMC Level 1 Scope

The CMMC Assessment Scope for Level 1 is defined in the official DoD CMMC Level 1 Scoping Guide. It covers every information system that processes, stores, or transmits FCI, along with the people, processes, and physical facilities that interact with those systems.

In practical terms, scope includes workstations and servers that handle FCI, cloud services used to store or transmit FCI, email systems used to send or receive FCI, file-sharing platforms holding FCI documents, network infrastructure carrying FCI traffic, physical facilities where any of the above are located, and personnel with access to any of the above.

Anything that does not touch FCI is out of scope. This is the simplest scoping model in CMMC, and it is also where most contractors trip up. The temptation is to declare a narrow scope (“just the one folder on the file server”) and ignore the email, the laptops, and the backups. Auditors and primes will not accept it.

Reach SOC 2 Compliance in 6 Weeks or Less

Schedule Your Free SOC 2 Assessment Today

CMMC Level 1 Requirements: All 15 Practices Explained

The 15 practices fall across six domains. Each is mapped to a NIST SP 800-171 control identifier, but Level 1 only assesses the subset of objectives relevant to FCI.

Access Control (AC)

AC.L1-B.1.I – Authorized Access Control

Practice: Limit information system access to authorized users, processes acting on behalf of authorized users, or devices.

Maintain a current list of users, processes, and devices authorized to access systems holding FCI. This means active user-account management: unique identifiers for each user, accounts disabled promptly when employment ends, and a documented process for reviewing who has access and why. Shared credentials are not acceptable. This is the foundation every other access control practice is built on, and it is where many contractors have their first reckoning with how loosely their environments have actually been managed.

AC.L1-B.1.II – Transaction and Function Control

Practice: Limit information system access to the types of transactions and functions that authorized users are permitted to execute.

Apply the principle of least privilege. A user with access to read FCI does not automatically get access to delete it, share it externally, or modify system configurations. Role-based access controls (RBAC) satisfy this requirement. In practice, this means auditing what each role can actually do in your systems and trimming permissions down to what is genuinely necessary for the job function.

AC.L1-B.1.III – External Connections

Practice: Verify and control or limit connections to and use of external information systems.

Know what external systems your in-scope environment connects to — cloud storage, partner networks, contractor laptops on home Wi-Fi — and apply controls to those connections. Acceptable Use Policies, VPN requirements, and explicit allow-lists for external sharing all map here. The key word is verify: you need documented evidence that external connections are inventoried and controlled, not just assumed to be fine.

AC.L1-B.1.IV – Control Public Information

Practice: Control information posted or processed on publicly accessible information systems.

Make sure FCI does not end up on your public website, your company blog, or any other publicly accessible system. This is mostly a process control: establish who is allowed to publish to public-facing systems and what review happens before anything goes live. It sounds obvious, but incidents involving inadvertent FCI disclosure through company websites and public repositories are more common than the industry likes to admit.

Identification and Authentication (IA)

IA.L1-B.1.V – Identification

Practice: Identify information system users, processes acting on behalf of users, or devices.

Every user, service account, and device that accesses FCI must have a unique identifier. Shared accounts — the classic “admin” login that three people use — are not acceptable. This applies to human users, automated processes, and devices alike. Document your user inventory and tie every access event to a specific, identifiable entity.

IA.L1-B.1.VI – Authentication

Practice: Authenticate (or verify) the identities of those users, processes, or devices as a prerequisite to allowing access to organizational information systems.

Passwords, multi-factor authentication, certificates, or biometric controls. Level 1 does not mandate MFA the way Level 2 does, but most modern environments implement it as the practical default — and assessors will note when they see environments that do not. Password complexity requirements, account lockout policies, and password reuse restrictions all live under this practice. For a deeper look at how authentication requirements scale across CMMC levels, the CMMC encryption requirements guide covers related technical controls in detail.

Media Protection (MP)

MP.L1-B.1.VII – Media Disposal

Practice: Sanitize or destroy information system media containing FCI before disposal or release for reuse.

When you decommission a hard drive, a USB stick, or a printer with internal storage, you must wipe or destroy it before it leaves your control. NIST SP 800-88 sanitization procedures define what “wiped” means in practice: a quick format is not enough for most media types. Throwing old laptops in a donation bin or the dumpster is an explicit failure of this control, and it is one of the most common findings in Level 1 assessments.

Physical Protection (PE)

PE.L1-B.1.VIII – Limit Physical Access

Practice: Limit physical access to organizational information systems, equipment, and the respective operating environments to authorized individuals.

Locked server rooms, badge-access offices, and clean-desk policies. If anyone can walk into your office and sit down at an unlocked workstation holding FCI, you fail this control. Physical access controls need to be commensurate with the environment: a home-based contractor and a 200-person manufacturer have different practical implementations, but the principle is the same.

PE.L1-B.1.IX – Manage Visitors and Physical Access

Practice: Escort visitors and monitor visitor activity; maintain audit logs of physical access; and control and manage physical access devices.

This practice combines visitor escort procedures, physical access logging, and management of access devices (badges, keys, smart cards) into a single practice with three assessment objectives. Visitor logs need to be real and current. Badge management needs to include a process for deactivating credentials when employees leave or access needs change.

Worth Knowing: The consolidation of PE.L1-B.1.IX

The consolidation of PE.L1-B.1.IX from three practices into one is why you will see both "17 practices" and "15 practices" in older guidance. The final 32 CFR Part 170 rule confirms the count is 15 practices with 17 assessment objectives. Vendors and consultants who still quote 17 practices are working from outdated drafts.

Reach SOC 2 Compliance in 6 Weeks or Less

Schedule Your Free SOC 2 Assessment Today

System and Communications Protection (SC)

SC.L1-B.1.X – Boundary Protection

Practice: Monitor, control, and protect organizational communications at the external boundaries and key internal boundaries of the information systems.

Firewalls at the network perimeter, network segmentation between trust zones, and monitoring of traffic at boundary points. For a small business, this can be as simple as a properly configured firewall with explicit allow rules and logged traffic. The key requirement is that the boundary is defined, protected, and monitored — not just assumed to exist because you have a router.

SC.L1-B.1.XI – Public-Access System Separation

Practice: Implement subnetworks for publicly accessible system components that are physically or logically separated from internal networks.

Your public-facing web server should not sit on the same network as your internal file server. Use a DMZ, separate VLANs, or hosted cloud separation to ensure that a compromise of a public system does not give an attacker a direct path into FCI-handling systems. If you have no publicly accessible systems within your assessment scope, this practice can be marked Not Applicable — but document that determination carefully.

System and Information Integrity (SI)

SI.L1-B.1.XII – Flaw Remediation

Practice: Identify, report, and correct information and information system flaws in a timely manner.

Patch management. Operating systems, applications, firmware — all of it. “Timely” is not defined with hard SLA language at Level 1, but assessors expect a documented patching process and evidence that patches are actually being applied, not just scheduled. A two-year-old critical vulnerability sitting unpatched is a finding regardless of what your policy document says.

SI.L1-B.1.XIII – Malicious Code Protection

Practice: Provide protection from malicious code at appropriate locations within organizational information systems.

Endpoint antivirus or EDR, email gateway scanning, and protection at any point where files enter the environment. Cloud-based protection counts, provided it is actually deployed and active on in-scope assets. “We have a license” is not the same as “it is installed and running on every covered endpoint.”

SI.L1-B.1.XIV – Update Malicious Code Protection

Practice: Update malicious code protection mechanisms when new releases are available.

Automatic signature updates and engine updates. The control is about currency, not just deployment. An antivirus product with definitions six months out of date fails this practice, even if the software itself is installed and running. Verify that automatic updates are enabled and confirm through configuration evidence — not assumption.

SI.L1-B.1.XV – System and File Scanning

Practice: Perform periodic scans of the information system and real-time scans of files from external sources as files are downloaded, opened, or executed.

Scheduled full-system scans plus real-time on-access scanning. Most commercial antivirus and EDR products do both by default, but you need evidence that the configuration is enforced across all in-scope endpoints — not just on the machine the IT manager happens to use. Export your endpoint management console settings. That is your evidence.

CMMC Level 1 Compliance Checklist

CMMC Level 1 Compliance Checklist

Before you begin a self-assessment, work through the following steps. This will not certify you, but it will surface most of the problems contractors encounter on their first attempt.

Confirm FCI handling. Identify every contract, vendor relationship, and data flow involving FCI. Review your DoD contracts and subcontracts for FAR 52.204-21 clauses.

Define your assessment scope. Document every system, location, and person that touches FCI. Reference the official CMMC Level 1 Scoping Guide and be prepared to defend every boundary decision.

Inventory your assets. Workstations, servers, cloud accounts, mobile devices, network gear. If it touches FCI, it is in scope and it needs to be listed.

Map controls to systems. For each of the 15 practices, document which technical or administrative measure satisfies it on each in-scope system. The mapping needs to be explicit, not assumed.

Identify your Affirming Official. This must be a senior representative with authority to bind the organization — a CEO, president, or designated corporate officer. An IT manager is not sufficient.

Gather evidence. Screenshots, configuration exports, policy documents, training records, vendor attestations. Evidence must exist before the assessment, not be created afterward.

Run a gap assessment. Self-test each practice. Since no POA&Ms are allowed at Level 1, anything not fully met must be remediated before the formal self-assessment.

Document a basic System Security Plan. Level 1 does not formally require an SSP, but assessors and primes will ask for one, and it becomes mandatory the moment you advance to Level 2. Start it now.

Register for SPRS access. Your Affirming Official needs a Procurement Integrated Enterprise Environment (PIEE) account with the SPRS Cyber Vendor User role.

Submit results and affirmation. Enter your Level 1 self-assessment results in SPRS and have the Affirming Official sign the attestation.

CMMC Level 1 Artifact Retention Requirements

Under 32 CFR 170.15(c)(2), the artifacts used as evidence for your Level 1 self-assessment must be retained for six years from the CMMC Status Date — the date the assessment was completed and entered in SPRS.

Six years matters because the DoD or a prime can ask to inspect your evidence at any point in that window. Refusing or being unable to produce it can trigger contract remedies, including suspension or debarment. The retention requirement also intersects with False Claims Act liability: if the affirmation submitted to SPRS turns out to have been inaccurate, the evidence chain is what gets investigated.

Practical retention covers the completed self-assessment record (objective-by-objective), all evidence cited in MET determinations, the senior-official affirmation signed in SPRS, any supporting policies, screenshots, and configuration exports referenced, and records of the assessment methodology used.

Most contractors store this in a dedicated compliance repository with version control.

Anything less invites a scramble during the next assessment cycle, since some evidence (like firewall rule snapshots) changes constantly and is hard to reconstruct after the fact.

Phased Implementation Timeline and Key Deadlines

CMMC implementation is rolling out in four phases under 32 CFR 170.3(e), on a timeline tied to the publication of the 48 CFR Acquisition Rule.

December 16, 2024: The 32 CFR Part 170 final rule takes effect, codifying CMMC as a federal regulation.

September 10, 2025: The 48 CFR DFARS final rule is published, amending DFARS 252.204-7021 to embed CMMC contract clauses.

November 10, 2025: Phase 1 begins. DoD solicitations can include CMMC Level 1 and Level 2 self-assessment requirements. This is the live enforcement date.

November 10, 2026: Phase 2 begins. Level 2 certification assessments (C3PAO-led) appear in more contracts at DoD discretion.

November 10, 2027: Phase 3 begins. Level 3 certification assessments enter contracts for the most sensitive CUI.

Full implementation is expected approximately three years after Phase 1, at which point all applicable contracts will contain CMMC clauses.

For Level 1 contractors, the operative date is November 10, 2025. From that point forward, a current Level 1 status in SPRS is a precondition for award on covered contracts. You do not get to fix it after the solicitation drops.

CMMC Level 1 Certification Challenges

Most Level 1 failures come from a small set of recurring problems.

Scope creep and scope denial. Contractors either underscope (excluding email, mobile devices, or cloud storage that clearly handles FCI) or overscope (sweeping in systems that have no FCI exposure). Both create problems. Underscoping risks false attestation. Overscoping wastes resources and creates evidence gaps that are hard to defend.

Missing evidence for inherited controls. Many Level 1 environments rely on cloud services (Microsoft 365, Google Workspace, AWS) to satisfy several practices. Inheritance is legitimate, but you need attestations from the provider, and you need to know which controls are inherited versus customer-responsibility. The shared responsibility model is real, and assessors check it.

Documentation that does not match reality. Policies that say one thing while configurations do another. The policy says workstations lock after 15 minutes; the actual Group Policy locks them after 60. Auditors compare written claims to system reality, and discrepancies become NOT MET findings.

Affirming Official confusion. The Affirming Official must be a senior representative with authority to bind the organization. An IT manager is typically not sufficient. A CEO, president, or designated corporate officer is the expected level. Some contractors discover this late in the process when no one in the org chart has the required authority formally documented.

Treating Level 1 as a checkbox. Level 1 is light compared to Level 2, but it is not nothing. The senior-official affirmation creates personal legal exposure, and the False Claims Act applies to false attestations the same way it applies to fraudulent invoices.

Important: The Department of Justice has been explicit that cybersecurity-related False Claims Act cases are an active enforcement priority. A signed CMMC affirmation that overstates compliance is precisely the kind of attestation that has produced multi-million-dollar settlements at other federal agencies. Take the senior-official signature seriously.

What Are the 15 CMMC Level 1 Requirements?

The 15 requirements come from FAR 52.204-21(b)(1) and cover Access Control (4 practices), Identification and Authentication (2), Media Protection (1), Physical Protection (2), System and Communications Protection (2), and System and Information Integrity (4). All 15 practices are described in full in the requirements section above.

Level 1 is self-assessment only. Contractors evaluate their own implementation and submit results to SPRS along with a senior-official affirmation. No C3PAO involvement is required at Level 1. Levels 2 and 3 involve third-party or government assessments depending on the contract type and data sensitivity.

Annually. The self-assessment must be performed every year, and a fresh senior-official affirmation must be submitted to SPRS within 12 months of the prior affirmation. Missing the deadline invalidates your status and makes you ineligible for new covered contract awards.

Noncompliance means loss of eligibility for new DoD contract awards requiring Level 1, potential suspension or termination of existing contracts, exclusion from prime contractor supply chains, and False Claims Act liability for any false attestations already submitted to SPRS.

For a small business with reasonable IT hygiene already in place, two to four weeks of focused effort is typical. For contractors starting from a less mature baseline, two to three months is more realistic. The bottleneck is usually evidence collection and scoping, not technical remediation. A structured gap assessment at the start of the process will give you a realistic timeline for your specific environment.

Yes. CMMC requirements flow down through the supply chain. Any subcontractor handling FCI in performance of a DoD subcontract needs its own current Level 1 status in SPRS. Primes are increasingly requiring proof of CMMC status before subcontract award, and this trend accelerated through 2025.

CMMC Level 1 covers 15 practices drawn from FAR 52.204-21 and addresses FCI only. NIST SP 800-171 contains 110 controls and addresses CUI, mapping to CMMC Level 2. The Level 1 practices are a subset of the broader 800-171 framework, but the two are not equivalent. A contractor compliant with NIST 800-171 is also compliant with Level 1; the reverse is not true. For a detailed breakdown of how the two frameworks compare, see our guide on CMMC vs NIST 800-171.

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.

Blog Highlights

Explore More Articles

On November 10, 2026, third-party certification becomes mandatory for most small defense contractors that handle Controlled Unclassified Information. That date, the start of Phase 2 of the CMMC rollout, is the one to circle in red.  The framework itself has been binding since the 32 CFR program rule took effect on December 16, 2024, and certification clauses began appearing in new contracts on November 10, 2025. Roughly 73 percent of the Defense Industrial Base (DIB) is made up of small businesses, and a 20-person machine shop now faces the same control set as a prime with a dedicated security team. This guide breaks down what the CMMC requirements for small business actually demand: the levels, the controls, the documentation, the real cost, and the route to certification. What Is CMMC and Who Needs to Comply? The Cybersecurity Maturity Model Certification is the Department of Defense’s program for verifying that contractors protect sensitive federal information on their own systems. For years, contractors simply self-attested compliance with NIST Special Publication 800-171. CMMC ends the honor system. It keeps self-assessment for lower-risk work and adds independent audits for everything else. Two regulations run the program. 32 CFR Part 170 defines the structure, the three levels, and the assessment rules. 48 CFR amends the Defense Federal Acquisition Regulation Supplement and embeds CMMC into contracts through clause DFARS 252.204-7021. The first sets the standard, the second makes it a condition of the award. Compliance is not optional based on company size. If you process, store, or transmit Federal Contract Information (FCI) or Controlled Unclassified Information (CUI) in the performance of a DoD contract or subcontract, CMMC applies. The requirement flows down from prime contractors to subcontractors and suppliers at every tier. The main carve-out is for companies that supply only commercially available off-the-shelf (COTS) products. The distinction between the two data types drives everything. FCI is information not meant for public release that is provided by or generated for the government under a contract. CUI is more sensitive: technical drawings, specifications, and procurement data the government requires you to safeguard. Which one you handle sets your level. The fastest way to check is your contract itself. Clauses such as DFARS 252.204-7012, 7019, and 7020 are strong signals that CUI is in scope. CMMC Levels Explained for Small Businesses CMMC has three levels. Most small contractors land at Level 1 or Level 2. Level 3 is reserved for a tiny fraction of the supply chain handling the most sensitive programs. Level 1 covers basic safeguarding of FCI. It maps to the 15 requirements in FAR 52.204-21, things most businesses already do, like using passwords and limiting who can access systems. Self-assessment is permitted and the cost is modest. Level 2 is where the majority of CUI-handling contractors sit. It requires all 110 security requirements in NIST SP 800-171 Revision 2, organized across 14 control families. Some non-prioritized contracts allow an annual self-assessment, but DoD estimates that around 95 percent of Level 2 contractors handle CUI critical enough to require a C3PAO assessment. Level 3 adds 24 selected enhanced requirements from NIST SP 800-172 on top of the full 110, for high-value programs targeted by advanced persistent threats. Assessments are conducted by the government’s Defense Industrial Base Cybersecurity Assessment Center (DIBCAC), and a contractor must already hold Level 2 certification before a Level 3 assessment can begin. Fewer than 1 percent of contractors will need it. To determine your level, read the solicitation and ask your prime directly. If any data you touch is CUI, plan for Level 2 and assume a third-party assessment until a contract tells you otherwise. Prepare Your Business for CMMC Compliance Get ready for the November 2026 CMMC deadline with expert guidance. Talk to a CMMC Expert Core CMMC Requirements Small Businesses Must Meet Level 2 requirements break into 14 control families covering 110 individual requirements and 320 assessment objectives. Access Control and System and Communications Protection are the two heaviest domains. In practice, these families fall into two buckets. The technical controls govern how your systems behave: limiting access to authorized users, requiring multi-factor authentication, logging system activity, hardening configurations, encrypting data, and detecting and responding to incidents. The administrative controls govern how your organization behaves: training staff, screening personnel, controlling physical spaces, assessing risk, and documenting everything. A small business cannot skip a family because it is inconvenient. There is no partial credit and no small-business exemption from the 110. Documentation Requirements for Small Business Compliance Assessors evaluate evidence, not intentions. Two documents anchor the entire effort. The System Security Plan (SSP) describes your environment, your CUI boundary, and how you implement each of the 110 controls. It is a living document and the first thing any assessor reads. The Plan of Action and Milestones (POA&M) records gaps, owners, and timelines for fixing them. CMMC scores Level 2 on a 110-point scale weighted by control importance. A score of at least 88 (80 percent) can earn conditional status, but only if certain high-value controls are fully met. Conditional status gives you 180 days to close every remaining item on your POA&M and pass a closeout assessment. Some critical controls cannot be deferred to a POA&M at all. Beyond the SSP and POA&M, you need written policies and procedures for each domain, plus concrete evidence that controls operate as documented: configuration screenshots, training logs, access reviews, and audit records. Pro Tip: Build your SSP Build your SSP before you spend a dollar on tools. Mapping your current state against all 110 requirements first tells you exactly where the gaps are, so you remediate the right things in the right order instead of buying software you may not need. Technical Requirements and Controls A handful of technical controls account for most assessment failures, and they deserve direct attention. The most effective cost and risk lever is scoping: isolating CUI into a dedicated enclave, a defined set of systems and networks, so the 110 controls apply only there rather than across your

Most security certifications were built for software that follows rules. AI agents do not. They consume data, draw conclusions, call tools, and take action, increasingly without a human in the loop. That gap is what AIUC-1 was created to close: it is the first auditable security standard built specifically for AI agents, and a few enterprise buyers have started asking vendors for it by name. This guide covers what AIUC-1 actually tests, the six risk domains it audits, how the certification process works, what it costs, how long it lasts, and how it aligns with SOC 2, ISO 42001, ISO 27001, and the NIST AI Risk Management Framework. It also covers the structural questions worth asking before you treat an AIUC-1 report as proof of anything. What Is AIUC-1 Certification? AIUC-1 is a certifiable standard for AI agents created by the Artificial Intelligence Underwriting Company (AIUC), a San Francisco-based, venture-backed startup founded by people with experience at organizations including Anthropic. The standard was developed with input from Orrick, Stanford, the Cloud Security Alliance, MIT, and MITRE, and launched in mid-2025. The framework comprises 51 requirements and 130 controls, organized across six risk pillars. It evaluates whether an organization has implemented and tested the technical guardrails, operational practices, and legal policies needed to reduce the risk of unsafe, unreliable, or unauthorized AI behavior. Certification applies to a specific AI system or product, not to the organization as a whole. An AIUC-1 certificate, audit report, and badge tell enterprise buyers that an agent has been independently tested against agent-specific risks. People describe AIUC-1 as the “SOC 2 for AI agents,” and the analogy holds in spirit. The difference is what it looks at. SOC 2 examines a service organization’s general controls. AIUC-1 examines how an agent behaves under pressure: when someone tries to jailbreak it, when it is asked to do something outside its scope, when it has access to data it should not expose. Worth Knowing: About AIUC-1 AIUC-1 does not define what counts as an “AI agent.” The vendor decides which system to certify and what falls in scope. That makes scope the single most important thing to check on any certificate, because a narrowly scoped audit may not cover the agent you actually use. Why AIUC-1 Certification Matters for Enterprise AI Adoption The business case rests on a simple problem: enterprises cannot reliably assess the security of their AI vendors, and the failures are expensive. According to EY research on responsible AI, 64% of companies with over $1 billion in revenue have already lost more than $1 million to AI-related failures.  That gap shows up directly in sales cycles. When security, legal, and procurement teams evaluate an AI vendor, they ask about hallucinations, prompt injection defenses, and what happens when an agent makes an unauthorized call. SOC 2 and ISO 27001 do not answer those questions. AIUC-1 gives buyers a structured, third-party-tested answer, which is why holding the certificate can move a stalled procurement review forward. The certification also produces real engineering outcomes, not just a badge. AIUC has reported cases where a customer service agent’s hallucination rate dropped from 11% to under 2% after strengthening its groundedness filter, and another where inappropriate-tone outputs fell from 9% to under 2% through better defensive prompting and output moderation. One company found and patched a PII exposure vulnerability during the certification process itself. The Six Core Risk Domains Covered by AIUC-1 AIUC-1’s 51 requirements are grouped into six domains. Each targets a category of risk that traditional security frameworks were not designed to handle. Data and Privacy Covers how customer data is used, retained, and protected. Requirements address input and output data policies, limits on what data the agent can access, protection of IP and trade secrets, prevention of cross-customer data exposure, and prevention of PII leakage. This is where the standard forces clarity on whether customer data trains the model and how long it is kept. Security The adversarial-resistance domain. It covers third-party testing of adversarial robustness, detection and real-time filtering of malicious inputs, prevention of prompt injection and unauthorized agent actions, enforcement of user access privileges, and protection of the deployment environment. This is the heart of what separates an agent audit from a general security audit. Safety Focuses on preventing harmful and out-of-scope outputs. Requirements include defining an AI risk taxonomy, conducting pre-deployment testing, preventing harmful and customer-defined high-risk outputs, and flagging high-risk outputs for human review. Safety is partly judgment-based, which means documentation alone can sometimes satisfy a requirement, so the testing behind it deserves scrutiny. Reliability Targets the failure modes that erode trust in production: hallucinations and tool misuse. Controls cover hallucination prevention and restrictions on which tools an agent can call and when. For a customer-facing agent, this is the domain that keeps it from inventing a refund policy or triggering the wrong workflow. Accountability Covers what happens when things go wrong. Requirements include AI failure response plans, vendor due diligence, and clear AI disclosure so users know when they are interacting with an agent. With human workers, accountability is built into org charts and chains of command. Agents need an equivalent, and this domain supplies it. Society The broadest domain, focused on preventing misuse with wider consequences: AI-enabled cyber attacks and CBRN (chemical, biological, radiological, nuclear) misuse. Most enterprise agents will touch only a few of these controls, but they matter for higher-capability systems. Insider Note: Of the 130 total controls, roughly 65 are mandatory, and 65 are optional. A straightforward agent typically needs to meet around 40 controls. A complex, multi-modal agent gets closer to 65. The scoping exercise determines which apply, so two AIUC-1 certificates can represent very different amounts of work. Ready to Earn Your AIUC-1 Certification? Accelerate Your AI Certification Journey Talk to an Expert Who Needs AIUC-1 Certification? AIUC-1 is built for any company developing or deploying agentic AI that sells into enterprises. The strongest fit is an organization whose product uses AI agents in customer-facing operations, handles

Most teams walk into a SOC 2 audit expecting standard requirements for their password policy: minimum length, 90-day rotation, one uppercase letter, one symbol, and so on. But there is no such checklist. The AICPA never published a list of mandatory password rules, and the federal guidance that most auditors lean on has thrown out half of what passed for best practice a decade ago.  Beyond compliance, this is remains a crucial cybersecurity control: Stolen and brute-forced credentials still drive a large share of breaches, and password policies are the main way to mitigate this risk. This guide covers what SOC 2 expects around passwords, where those expectations come from, and how to build a policy that satisfies an auditor without making your security worse. What Are SOC 2 Password Requirements? SOC 2 password requirements are the access controls that a service organization implements to govern how passwords are created, stored, enforced, and retired, all in service of the Trust Services Criteria. The important word is controls, not rules. SOC 2 does not hand you a specification. It asks whether your controls are suitably designed and operating effectively to keep unauthorized people out of your systems.   The Role of Passwords in the SOC 2 Trust Services Criteria The Trust Services Criteria, developed by the AICPA, are the evaluation standard for every SOC 2 report. Passwords sit inside the Security category, which is mandatory in all SOC 2 engagements, and specifically inside the Common Criteria series CC6, covering logical and physical access. Passwords are one of the most basic logical access controls you have, and one of the most scrutinized, because CC6 is usually the most evidence-intensive part of the entire audit. Relevant Common Criteria: CC6.1, CC6.2, and CC6.3 CC6.1 covers the controls that restrict logical access to systems, infrastructure, and data, this is where your password policy, MFA enforcement, and account lockout settings live. CC6.2 governs how access is granted, modified, and removed, meaning your provisioning workflows, access reviews, and offboarding processes are all evaluated here. CC6.3 focuses on the removal of access when it is no longer needed and the management of privileged credentials specifically. Together, these three criteria map to the full lifecycle of a credential: creation, ongoing use, and retirement. An auditor working through CC6 will expect evidence at every stage.   Does SOC 2 Mandate Specific Password Rules? No. The AICPA is explicit that the Trust Services Criteria do not define the controls an organization must have. You identify and implement controls that meet the criteria, and the auditor evaluates them. That means there is no AICPA-mandated minimum length, no required rotation interval, and no prescribed complexity formula. What the auditor checks is whether your stated controls exist, work, and reasonably prevent unauthorized access. Insider note: Auditors rarely fail you for choosing a 10-character minimum over 12. They fail you when your written policy says one thing and your actual system configuration says another. Consistency between the policy document and the enforced setting matters far more than the specific number. Why Password Requirements Matter for SOC 2 Compliance Preventing Unauthorized Access Credentials are the front door. The 2025 Verizon DBIR found that stolen credentials remained the single most common initial access vector, appearing in 22% of breaches, and that brute force attacks against basic web applications nearly tripled year over year. Strong authentication controls are the difference between an attacker hitting a wall and an attacker walking straight in with a valid login. Reducing Data Breach Risk Weak or reused passwords feed credential stuffing, where attackers replay username and password pairs harvested from earlier breaches against your login pages. Reuse is rampant: research from Microsoft’s Digital Defense Report routinely finds that the majority of people reuse passwords across services. A single leaked password elsewhere becomes a working key to your environment unless your controls catch it. Demonstrating Logical Access Controls to Auditors SOC 2 is an attestation. It is not enough to be secure; you have to prove it with evidence. Well-designed password controls produce exactly the artifacts an auditor wants: configuration screenshots, enforcement logs, MFA reports, and access review records. Good controls and good evidence are two sides of the same coin, and an internal audit process that routinely collects this evidence makes the formal engagement significantly less stressful. Core SOC 2 Password Requirements Although SOC 2 prescribes nothing specific, a defensible password policy almost always addresses the same set of controls. These are what auditors expect to see and what your peers in compliance treat as table stakes. Minimum Password Length Length is the strongest single lever for password entropy, and modern guidance favors it over everything else. A common defensible baseline is at least 12 characters for standard user accounts, with longer requirements for service and admin accounts. NIST SP 800-63B recommends that verifiers support passwords up to 64 characters so that passphrases and password-manager output are never truncated, an important implementation detail that many teams overlook. Password Complexity and Blocklists Old-style complexity rules, one uppercase, one symbol, one number, are fading, and for good reason. They push users toward predictable substitutions without meaningfully raising entropy. The more effective control is a blocklist: screening new passwords against dictionaries of common and previously breached credentials and rejecting matches. Tools like Have I Been Pwned’s Pwned Passwords API make this straightforward to implement. This stops Password1! from sneaking through even though it technically satisfies a legacy complexity rule. Password Rotation and History Forced periodic rotation is the control most teams keep out of habit, and it is also the one that modern guidance most clearly discourages. Rotation pushes users toward predictable patterns, Spring2025 becoming Summer2025, without improving security in any measurable way. Password history settings, which prevent the immediate reuse of recent passwords, still have a place, but blind calendar-based expiry should be replaced with event-driven resets: force a change when there is evidence of compromise, not because the calendar says 90 days have passed. Account Lockout After Failed Login Attempts An account