Risk acceptance and delegated decisions: the act with a bearer
Accepting risk is a decision with a bearer, not the absence of a decision.

When an organisation decides to live with a risk it cannot or will not remove, that is a risk acceptance — a decision taken by someone and recorded somewhere, not what happens when nobody acts. NIST’s risk framework puts it plainly: an authorising official issues an authorisation “accepting the security and privacy risks to the organization’s operations and assets” (SP 800-37r2, lines 501–503). Acceptance is a named official assuming responsibility on the basis of a defined set of controls.
Because acceptance is an act, it can be delegated — and delegation is what makes it work across a large organisation. But delegation moves the authority to decide, not the consequence of being wrong. This sheet covers the act and the transfer: whose decision it is, the vocabulary that keeps it auditable, and the point at which it stops being one.
What acceptance actually is
The framework defines acceptance by its bearer. The authorising official “is the only organizational official who can accept the security and privacy risk to organizational operations, organizational assets, and individuals” (SP 800-37r2, lines 7316–7318), and holds “the authority to formally assume responsibility and accountability for operating a system” (lines 7314–7315). The act is explicit and located: “the final action before a system is placed into operation is the explicit acceptance of risk by the authorizing official” (lines 2013–2014). Acceptance is therefore someone’s — a role with a name — and it is attached to something: a system, or a defined set of controls.
decision rights the act what the delegation carries
--------------- ------- ---------------------------
who holds the authority -> ACCEPT risk -> a named risk owner
(how security enters on the basis of a the basis recorded
decisions: sheet 1) defined control set the tolerance stated
(residual, after assurance kept separate
the response step) (third line)
Acceptance comes after the response, not instead of it
Acceptance is one of a set of responses, chosen once the risk has been assessed. NIST SP 800-39 defines the step: “Risk response identifies, evaluates, decides on, and implements appropriate courses of action to accept, avoid, mitigate, share, or transfer risk” (lines 2750–2752). It then states what is left over: “Based on the course of action selected during the risk response step, some residual risk may remain. Under certain circumstances, the level of residual risk could trigger a reassessment of risk” (lines 2724–2726). Acceptance is a step inside the process, whose leftover risk can reopen it.
What is accepted is measured against an organisation-level limit: “The level of residual risk accepted by authorizing officials for information systems or inherited common controls is within the organizational risk tolerance, and not the individual risk tolerances of those authorizing officials” (lines 2463–2466). An acceptance is legitimate when it sits inside the tolerance the organisation set.
The ERM vocabulary that makes the delegation auditable
Enterprise risk management supplies the words that make a decision followable. NISTIR 8286r1 defines risk appetite as “the amount and type of risk that an organization is willing to take in meeting its objectives” (lines 581–583) and risk tolerance as “the organization or stakeholders’ readiness to bear the remaining risk after responding to or considering the risk to achieve its objectives” (lines 587–589). An acceptance with no tolerance to measure it against cannot be judged afterwards.
The same document places acceptance in a named sequence: “5. Plan and execute risk response strategies. The appropriate response is determined for each risk, and the decisions are informed by risk guidance from leadership” (lines 676–677). It names the hand-down too: enterprise decision-makers “use the risk profile to choose which enterprise risks to address, allocate resources, and delegate responsibilities to appropriate risk owners” (lines 355–357). Delegation is recorded — owner, resource, guidance — so the chain stays retrievable.
Where the law names the responsible body
Regulation does not leave the bearer to inference. DORA states the principle for financial entities: “the principle of the management body’s full and ultimate responsibility for the management of the ICT risk” (Recital 46, line 110). NIS2 makes the body the approver: “Member States shall ensure that the management bodies of essential and important entities approve the cybersecurity risk-management measures taken by those entities” (Article 20, line 845). The body can delegate the work of deciding, not the accountability for it.
Who decides is not who assures
Acceptance is only meaningful if the roles that decide are separated from the roles that check. The IIA’s Three Lines Model defines the separation: first line roles are “most directly aligned with the delivery of products and/or services”, second line roles “provide assistance with managing risk” (lines 129–130), while internal audit, the third line, “provides independent and objective assurance and advice on the adequacy and effectiveness of governance and risk management” (lines 142–144); the governing body, in the model, is the role that delegates responsibility to management and provides the resources for it (line 117). The official who accepts risk is a management role; the assurance that tests whether the acceptance was sound belongs to a role accountable to the governing body, not the acceptor.
Where an acceptance stops being one
Three limits follow, each a way a sheet of paper can look like a decision while being none. First, accepting risk without a stated tolerance, a named owner and an expiry is a silence, not a decision: nothing can be tested against it and nothing ends. Second, delegating the decision does not delegate the consequence — the authority to accept can move down, but the accountability the legal texts place on the management body does not move with it. Third, an acceptance nobody can retrieve when the risk materialises is not an acceptance but a memo; the vocabulary above exists so that the decision, its owner and its basis stay findable.
The boundary is deliberate: how security enters decisions is the companion sheet’s subject; the bridge from residual risk into strategy belongs to the metrics-and-strategy sheet; and the machinery of individual exceptions and approvals is the L3 sheet’s, pointed at here and not rebuilt.
Level and prerequisites
L5 — specialist: the sheet is about the acceptance act and the delegation that carries it across functions, not about operating a register or designing a control. Prerequisites: the L1 sheets on risk appetite and acceptance criteria and on risk treatment options; the L2 risk and exception register and the roles, responsibilities and RACI sheet; and the L3 sheet on exceptions, approvals and compensating controls, which owns the mechanics this sheet takes as given.
Where to go next
- Knowledge — the Knowledge register this sheet belongs to.
- Cybersecurity Governance — the macro-area.
- Risk management — the node this sheet sits in.
References
- NIST — SP 800-37 Rev. 2, Risk Management Framework — the acceptance of security and privacy risk, the authorising official as the only official who can accept it, and the explicit final acceptance before operation.
- NIST — SP 800-39, Managing Information Security Risk — the risk response step, the residual risk it leaves, the reassessment trigger, and acceptance within organisational risk tolerance.
- NIST — IR 8286r1, Integrating Cybersecurity and Enterprise Risk Management — risk appetite, risk tolerance, the risk-response-strategy step and the delegation to risk owners.
- DORA — Regulation (EU) 2022/2554 — the management body’s full and ultimate responsibility for ICT risk.
- NIS2 — Directive (EU) 2022/2555 — the management body’s approval of cyber risk-management measures and its liability.
- The IIA — The Three Lines Model — the separation of the roles that decide from the roles that assure.