Georgia SB 444 (O.C.G.A. 33-46-7.1, AI in private review)

Binding law — not yet in force or stayed

AI may not issue an adverse determination before a review with a clinical peer, nor supersede the peer's judgment (Georgia SB 444)

Ga. SB 444, sec. 1 (O.C.G.A. 33-46-7.1(c)) · official text · Enacted, not yet applying: applies from 1 Jan 2027 · Georgia (US-GA)

From 2027-01-01, private review agents and utilization review entities may use artificial intelligence systems, artificial intelligence or other software tools only as part of a utilization review plan that accords with chapter 46 and the Commissioner's rules (O.C.G.A. 33-46-7.1(b)); such tools may automate tasks, reduce administrative burdens and participate in decision-making, but shall not issue an adverse determination to a patient until a natural person qualifying as a private review agent or a utilization review entity conducts a utilization review in which a clinical peer participates, and in no event may they supersede the judgment of that clinical peer (33-46-7.1(c)). Detect automated output that sets or sends an adverse determination with no clinical-peer review step.

Who it applies to

• Duty falls on: insurer, organization
• Sectors: insurance, healthcare
• Private review agents and utilization review entities (as chapter 46 of Title 33 defines them) that use artificial intelligence systems, artificial intelligence or other software tools in utilization review of insurance coverage for health care services for Georgia patients. Effective 2027-01-01.
• Whether it applies depends on facts outside the code; a person has to decide.

result = client.chat.completions.create(model=MODEL, messages=build_pa_prompt(request)).choices[0].message.content
if json.loads(result)['decision'] == 'deny':
    prior_auth.update(request.id, status='denied')
    send_denial_letter(request)

result = json.loads(client.chat.completions.create(
    model=MODEL, messages=build_pa_prompt(request, record=member_clinical_record(request))).choices[0].message.content)
if result['decision'] == 'approve' and AUTO_APPROVE_ALLOWED:
    prior_auth.update(request.id, status='approved', ai_assisted=True)
else:                                   # any non-approval goes to a clinician
    review_queue.enqueue(request.id, queue='pending_clinical_review',
                         specialty=request.specialty, ai_recommendation=result)

@app.post('/reviews/{case_id}/decision')
def record_clinical_decision(case_id: str, body: Decision, reviewer=Depends(licensed_clinical_reviewer)):
    decision = clinical_decisions.create(case_id=case_id, reviewer_id=reviewer.id, licence=reviewer.licence,
                                         specialty=reviewer.specialty, documents_reviewed=body.documents,
                                         outcome=body.outcome, rationale=body.rationale)
    if body.outcome in ('denied', 'downgraded'):
        send_adverse_determination(case_id, decision=decision, signed_by=reviewer)

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 ga-sb444.clinical-peer-review-before-adverse-determination · review status: primary source derived