Binding law — not yet in force or stayed AI-adjacent law
Decisions from automated processing with legal or serious effects need an objection route and, on request, human review (UAE PDPL Art. 18)
The UAE's Federal Decree-Law No. 45 of 2021 on the Protection of Personal Data gives the data subject a right to object to decisions issued by automated processing that have legal consequences or seriously affect them, including profiling (Art. 18(1)). There is no objection where the automated processing is within the terms of a contract between the data subject and the controller, is required by other legislation in force in the State, or rests on the data subject's prior consent under Art. 6 (Art. 18(2)); in those cases the controller must still apply suitable measures to protect the data's privacy and confidentiality and the person's rights (Art. 18(3)). In every case the controller must bring the human element into the review of automated-processing decisions at the data subject's request (Art. 18(4)). 'Automated processing' covers processing by a program or electronic system working automatically, wholly without human intervention or partly with limited human supervision (Art. 1). Detect a model output that becomes a decision about a person with no objection route and no human review on request, and no human decision.
Trust and provenance not reviewed by a lawyer · audit-grade · source verified 3 Oct 2026 · release 2026.10.03.4
- Lane
- Binding law — not yet in force or stayed Application uncertain: dated 2 Jan 2022, enforcement status not confirmed; check the official source
- Official source
- Art. 18(1) (right to object to automated-processing decisions with legal or serious effects, including profiling) · captured 3 Oct 2026 · anchor hash (SHA-256)
0b022fe10d46…· 11 more anchors in the data release - Verification
- Quoted text found word for word in the captured official document (3 Oct 2026). Source last verified 3 Oct 2026: checked against the captured official document; not in the weekly watcher's list; checked against the captured document.
- Data release
- Data release 2026.10.03.4, data as of 3 Oct 2026, schema 0.3.10. This page also reflects corpus changes made after that release; they ship in the next one.
- Legal review
- Not reviewed by a lawyer. TwinEthos derived this rule from the official text it cites: treat it as research to check against that text; it is not legal advice. No TwinEthos rule has been legally reviewed yet. Open questions for counsel on this rule: 1.
- Audit standard
- Audit-grade: meets all 10 checks of the TwinEthos audit standard that apply to it. The audit standard is TwinEthos's own quality bar for provenance, dates, applicability, detectors, fixtures, remediation and licences; it is not a legal review.
- Detectors
1 detector (data flow), experimental: written from the rule's text and not yet measured for precision on real code, so treat a hit as a lead to verify.
Known limits:
- Whether a contract, other legislation or prior consent removes the objection right (Art. 18(2)) is not detected
- The privacy and confidentiality measures of Art. 18(3) are not detected
- Decisions without legal consequences or serious effects are outside Art. 18(1); the objection form may live in a separate service.
Who it applies to
- Duty falls on: controller
- Systems covered: automated decision, consequential decision
- Controllers in the UAE, and controllers outside the UAE processing personal data of data subjects in the UAE (Art. 2(1)), that issue decisions by automated processing, including profiling, with legal consequences or serious effects on the data subject. The Decree-Law is in force from 2022-01-02 (Art. 31); Art. 29 gives controllers six months from the issue of the Executive Regulation to adjust, and whether that Regulation has been issued, and so whether Art. 18 is enforced, is a counsel question. Excluded: government data and authorities, security and judicial authorities, personal use, health and banking data with their own legislation, and free zones with their own data protection laws (Art. 2(2)).
- Not covered:
- Government data; government authorities that control or process personal data; personal data held by security and judicial authorities (Art. 2(2)(a)-(c))
- A data subject processing their own data for personal purposes (Art. 2(2)(d))
- Health personal data and banking and credit personal data that have their own protection legislation (Art. 2(2)(e)-(f))
- Companies and establishments in the State's free zones that have their own personal data protection legislation, such as the DIFC and ADGM (Art. 2(2)(g))
- Whether it applies depends on facts outside the code; a person has to decide.
The guard to add
Route significant automated decisions through meaningful human review, or wire in an automated-decision notice, reasons, human intervention, a way to give a view, and contest.
At the point where model output becomes a significant decision about a person (approve, deny, underwrite, set_status), either queue the case for a reviewer who weighs the evidence and can change the outcome before it takes effect (review_queue.enqueue, requires_human_review), or, where the decision stays solely automated, record the permitted basis for that decision type and wire the safeguards in. Those safeguards are a notice in the decision message that it was made by automated processing, reasons the person can read, and request_human_review or contest routes where the person can give their view and have a human reconsider. A reviewer who approves every case without examining it does not make the decision non-automated, so the review records reviewer identity, the evidence viewed, and the outcome.
Where it goes: 1 application source code, 9 AI output handling, 15 agent action surface, 14 user-facing text.
What this provision adds:
- Give the person a way to object to a decision from automated processing that has legal consequences or seriously affects them, including profiling, unless an Art. 18(2) case applies.
- On the person's request, have a human review the automated decision, in every case (Art. 18(4)).
- Where the objection right is excluded by contract, other legislation or prior consent, keep suitable measures protecting the privacy and confidentiality of the data and the person's rights (Art. 18(3)).
Example (Python + OpenAI SDK), before:
verdict = client.chat.completions.create(model=MODEL, messages=msgs).choices[0].message.content
if verdict.strip() == 'deny':
deny(applicant)
send_decision_email(applicant, 'Your application was not approved.')After:
out = client.chat.completions.create(model=MODEL, messages=msgs,
response_format={'type': 'json_object'})
result = json.loads(out.choices[0].message.content)
if result['decision'] == 'deny':
if requires_human_review('credit'): # a person decides
review_queue.enqueue(applicant.id, proposal=result)
else: # solely automated, recorded basis
deny(applicant, basis=DECISION_BASIS['credit'], reasons=result['reasons'])
send_decision_email(applicant, render('adm_denial.txt', notice=ADM_NOTICE,
reasons=result['reasons'], contest_url=f'{BASE}/decisions/{applicant.id}/contest'))Control: Solely-automated significant decision without human-intervention safeguards. The same guard addresses 18 items with binding law in 18 jurisdictions. Engineering guidance, not legal advice.
Related incidents
No guardrail sits on this exact control; these incidents are cited by guardrails on related controls.
- UnitedHealth nH Predict claim-denial litigation (2023-11; alleged (not proven)). A class action filed in November 2023 alleges that UnitedHealth's nH Predict model had a 90% error rate, measured by denials reversed on appeal, while only about 0.2% of members appealed. UnitedHealth disputes the allegations; the litigation is ongoing. Source: STAT News · evidence grade: primary · cited by Monitor how often adverse AI decisions are reversed, and suspend models that are usually wrong
- Cigna PXDX batch claim denials (reported) (2022; alleged (not proven)). ProPublica, citing internal Cigna records, reported that Cigna's PXDX system was used to reject more than 300,000 claims over two months in 2022, with physicians spending an average of 1.2 seconds on each. Cigna disputes the reporting; related lawsuits are ongoing. Source: ProPublica / The Capitol Forum · evidence grade: press of record · cited by Make human review of adverse AI decisions substantive, not nominal
Rule id ae-pdpl-art18.object-and-human-review · review status: primary source derived