The Government Did Not Buy a Cybersecurity Promise. It Bought the Controls.
A cybersecurity requirement can look like technical boilerplate until the government attaches a dollar figure to it. On November 13, the Justice Department announced that Johnson Controls and two subsidiaries agreed to pay $5 million to resolve allegations that they failed to satisfy cybersecurity obligations in 25 federal contracts and subcontracts. The government alleged that the companies lacked sufficient knowledge and control of their information systems, did not maintain a contractually adequate System Security Plan for covered work, and did not adequately document, develop, or implement a plan to correct deficiencies. Johnson Controls did not admit liability, and the settlement is not a judicial finding that the allegations were true. Its practical message is still sharp. When a contract incorporates cybersecurity requirements, the government is not buying a policy statement or a future intention. It is buying operating controls, a current description of those controls, and an honest record of what remains unfinished.
This is a procurement case before it is a breach story
The settlement covers work for the Departments of Defense, Energy, and Justice, as well as NASA. The agreement says the 25 contracts involved protected information, including Controlled Unclassified Information, Official Use Only material, and other restricted categories. Different agencies used different clauses, but the common premise was simple: information entrusted to a contractor did not lose its required protection when it moved outside a government system.
The government's allegations focus on performance. It contended that Johnson Controls lacked sufficient information about its systems or sufficient control over them and therefore failed to restrict access as the contracts required. For contracts incorporating Defense Federal Acquisition Regulation Supplement clauses tied to NIST Special Publication 800-171, the government also alleged deficiencies in the System Security Plan and in the plan of action and milestones used to address unmet requirements.
That framing matters. Public cybersecurity discussions often begin with an intrusion, stolen data, or operational disruption. A contract case can begin earlier. The question is not only whether an attacker succeeded. It is whether the contractor supplied the safeguards, documentation, and remediation process the government paid to receive.
The distinction also protects against overstatement. The Defense Department's current policy says that reporting a cyber incident does not by itself prove a failure to provide adequate security. A breach is evidence to investigate, not automatic liability. The inverse is equally important. The absence of a publicly disclosed breach does not establish contractual compliance if the promised controls were missing or the records did not accurately describe the system.
The security plan is part map, part representation
NIST SP 800-171 supplies security requirements for protecting Controlled Unclassified Information in nonfederal systems. The relevant contract clauses required implementation of those safeguards. The settlement agreement describes what the System Security Plan was expected to do: identify system boundaries and operating environments, explain how requirements were implemented, and describe connections to other systems.
A useful plan therefore does more than announce that a company takes security seriously. It tells a reviewer which systems are inside the protected environment, where sensitive information moves, which controls operate, and how the company knows. If the boundary is wrong, the rest of the assessment may be wrong too. An overlooked cloud tenant, vendor portal, building-control network, or support account can sit outside the diagram while remaining inside the actual work.
The plan also functions as a representation about present conditions. That does not mean every drafting error is fraud. Systems change, inventories age, and reasonable professionals may disagree about scope. But a plan that remains static while the environment changes can become evidence that the organization stopped testing its written account against reality.
Lawyers should recognize the pattern. The document may be created by security personnel, but its legal significance depends on the contract, the claims for payment, the people who relied on it, and what the company knew when it spoke. Calling it a technical artifact does not keep it outside discovery, an inspector general inquiry, a whistleblower complaint, or a materiality analysis.
A remediation list can be honest or dangerous
Federal cybersecurity frameworks do not assume that every contractor satisfies every requirement at every moment. The plan of action and milestones exists because deficiencies occur. It identifies an unmet requirement, the work needed to correct it, the responsible party, and the expected schedule. Used honestly, it converts a gap into a managed obligation.
The legal risk appears when that record stops managing anything. A deficiency can remain open without a credible owner. A completion date can move repeatedly without explanation. A compensating control can be named but never tested. A milestone can be marked complete because a document changed while the system did not. At that point, the remediation file may tell a more damaging story than no file at all because it can show knowledge, duration, and the distance between the recorded response and actual performance.
The settlement agreement alleges that Johnson Controls did not adequately document, fully develop, or fully implement its remediation plan. That is the government's contention, not an adjudicated fact. The wording nevertheless shows why counsel should not reduce a plan of action to a list prepared for an assessor. It is a time-based record of what the company knew and what it said it would do.
Good governance separates acceptance of risk from silent drift. If a deadline changes, the record should explain why, identify interim protection, state who approved the decision, and set the next verification point. If the contract does not permit performance with an open gap, a detailed remediation plan cannot amend the bargain. It can surface the problem for a disclosure, waiver, contract modification, or operational decision before the next invoice turns uncertainty into a claim.
The False Claims Act is not a strict-liability security code
The Justice Department brought this resolution under the False Claims Act, but cybersecurity imperfection alone does not establish an FCA violation. Section 3729 requires a qualifying false or fraudulent claim, or a false statement material to such a claim, plus the required knowledge. Knowledge includes actual knowledge, deliberate ignorance, and reckless disregard. Materiality asks whether the matter had a natural tendency to influence, or was capable of influencing, payment or receipt of money.
Those elements matter because security standards involve judgment, inheritance, system boundaries, contract interpretation, and changing technology. A later assessor's disagreement does not automatically prove that a prior claim was knowingly false. Neither does a vulnerability prove that the contractor promised it could never be breached.
The sharper cases involve the relationship among the contract, the company's internal evidence, and payment. Did the clause apply to this system? What did the company represent about implementation? Who knew of the gap? Was the government told? Did the company continue submitting claims while treating a required control as optional? Would the information have mattered to the contracting decision or payment? Those questions turn a control failure into a legal theory, or show why the theory fails.
This is why compliance teams should resist both panic and complacency. Treating every open item as fraud produces noise and discourages candid security reporting. Treating every control as merely technical ignores the representations embedded in procurement. Counsel's job is to map the facts to the actual clause and statutory elements, not to convert one vocabulary into the other by slogan.
The whistleblower award changes who can see the case first
The settlement resolves two qui tam actions filed under seal by private relators. The agreement provides that one relator, former Johnson Controls employee Kelly Vance, will receive $850,000 from the government's recovery. It also provides separate payments for relators' counsel fees and costs. Those terms illustrate the economic design of the False Claims Act: insiders can bring claims on behalf of the United States and may share in a recovery.
That incentive changes the value of internal evidence. A security engineer, contract manager, auditor, or program employee may see the difference between the plan and the system before senior legal leadership does. If an internal report receives a narrow technical answer when it raises a contractual representation, the organization can miss the point of the warning.
This does not mean every disagreement should be treated as a threatened qui tam suit. It means reporting systems must route cybersecurity concerns across functions. Security can determine whether a control operates. Contracts can determine what the clause requires. Finance can identify the claims and payments involved. Counsel can evaluate knowledge, materiality, disclosure, privilege, and remediation. No single department owns the complete fact pattern.
DOJ launched its Civil Cyber-Fraud Initiative to use the False Claims Act against knowing cybersecurity misrepresentations and failures tied to federal funds. The Johnson Controls settlement is part of that continuing enforcement architecture. The lesson for contractors is not to promise perfection. It is to know what the contract bought, maintain evidence of what was delivered, make gaps visible, and tell the truth before an invoice or certification tells the story for you.
Sources and further reading
Primary and industry sources used to support this page. External guidance should be reviewed in context and for your jurisdiction.
- Justice Department, Johnson Controls cybersecurity settlementOfficial November 13, 2026 announcement of the $5 million resolution, the alleged contractual failures, the $850,000 relator share, and the statement that the allegations were not adjudicated.
- Johnson Controls settlement agreementPrimary agreement identifying the 25 contracts, covered information, alleged conduct, payment allocation, releases, reservations, and no-admission language.
- NIST SP 800-171 Revision 3Current NIST publication describing security requirements for Controlled Unclassified Information in nonfederal systems and organizations.
- DFARS 252.204-7012Official acquisition clause defining adequate security, covered defense information, covered contractor systems, and safeguarding obligations.
- 31 U.S.C. § 3729Current False Claims Act text governing false claims, false statements, knowledge, materiality, obligations, and remedies.
- 31 U.S.C. § 3730Current statutory provisions governing civil actions, government intervention, private relators, awards, and related procedure.
- Justice Department Civil Cyber-Fraud InitiativeDOJ's official explanation of its use of the False Claims Act for knowing cybersecurity misrepresentations and contractual failures.