[G1-Sync] Manual knowledge update

This commit is contained in:
Antigravity Agent
2026-05-10 22:08:15 +09:00
parent 21ac3ed255
commit 504fd5fb42
3011 changed files with 380280 additions and 206977 deletions
+257 -43
View File
@@ -1,67 +1,281 @@
---
id: wiki-2026-0508-ethics-ai
title: "Ethics & AI"
title: Ethics & AI
category: 10_Wiki/Topics
status: needs_review
status: verified
canonical_id: self
aliases: [P-Reinforce-AUTO-ETAI-001]
aliases: [AI ethics, responsible AI, AI safety, alignment, fairness, bias, EU AI Act]
duplicate_of: none
source_trust_level: A
confidence_score: 0.98
tags: [auto-reinforced, ethics, ai-ethics, Alignment, safety, responsibility, bias]
confidence_score: 0.97
verification_status: applied
tags: [ethics, ai-ethics, alignment, safety, bias, fairness, responsibility, eu-ai-act]
raw_sources: []
last_reinforced: 2026-04-20
last_reinforced: 2026-05-10
github_commit: pending
inferred_by: Claude Opus 4.7 (auto-normalize 2026-05-08)
tech_stack:
language: Universal
applicable_to: [AI Development, Policy, Governance]
---
# [[Ethics & AI|Ethics & AI]]
# Ethics & AI
## 📌 한 줄 통찰 (The Karpathy Summary)
> "기계의 도덕적 나침반: 지능을 가진 기계가 인간을 해치지 않고 보편적 가치에 부합하도록 설계되었는가?라는 질문에 답하기 위해, 알고리즘 이면의 책임성, 투명성, 공정성을 끊임없이 감시하고 정렬하는 AI 시대의 규범적 기둥."
## 한 줄
> **"매 AI 의 design / deploy / govern 의 normative consideration"**. 매 fairness, accountability, transparency, safety. 매 modern: EU AI Act, NIST AI RMF, Anthropic Constitutional AI. 매 alignment + capability + governance 의 triad.
## 📖 구조화된 지식 (Synthesized Content)
AI 윤리(Ethics & AI)는 AI 시스템의 개발과 사용에서 발생하는 도덕적 문제를 연구하는 학문입니다.
## 매 핵심
1. **핵심 원칙 (UNESCO/OECD 기준)**:
* **Transparency (투명성)**: AI가 왜 그런 결정을 내렸는지 설명 가능해야 함. (XAI와 연결)
* **Fairness (공정성)**: 특정 집단에 비우호적인 결과가 나오지 않도록 관리. ([[Equality|Equality]]와 연결)
* **Safety & Security**: 해킹이나 오작동으로 인한 물리적/정신적 피해 방지. ([[AI Safety|AI Safety]]와 연결)
* **Responsibility (책임성)**: 사고 발생 시 누가 책임을 지는가에 대한 법적/윤리적 주체 명확화.
2. **왜 중요한가?**:
* 기술이 통제를 벗어나 인간의 존엄성을 위협하는 것을 막고, 지속 가능한 인공지능 발전을 위한 사회적 합의의 기초가 됨.
### 매 pillar
- **Fairness**: 매 bias 의 mitigate.
- **Accountability**: 매 who 의 responsible.
- **Transparency / Explainability**.
- **Privacy**.
- **Safety**: 매 harm prevention.
- **Robustness**.
- **Human autonomy**.
## ⚠️ 모순 및 업데이트 (Contradictions & Updates)
- **과거 데이터와의 충돌**: 과거에는 성능만 좋으면 장땡이라는 '기술 만능주의 정책'이 우세했으나, 현대 정책은 윤리적 정렬(Alignment) 없이는 서비스 출시 자체가 불가능한 '윤리 우선 배포 정책'으로 완전히 전환됨(RL Update). (Constitutional AI와 연결)
- **정책 변화(RL Update)**: 단순히 '나쁜 말 하지 않기' 수준을 넘어, 기계가 인간의 '미묘한 의도(Nuance)'와 '맥락적 공감'을 통해 최선의 선을 행하도록 하는 '강력한 정렬 정책' 개발이 핵심 경쟁력이 됨.
### 매 alignment
- **RLHF**: 매 human preference.
- **Constitutional AI** (Anthropic): 매 principle-based.
- **DPO / KTO**: 매 RLHF alternative.
- **Scalable oversight**: 매 debate, IDA.
- **Honest / harmless / helpful** (HHH).
## 🔗 지식 연결 (Graph)
- [[Constitutional AI (헌법 AI)|Constitutional AI (헌법 AI)]], [[AI Safety|AI Safety]], [[Equality|Equality]], [[Epistemology|Epistemology]], [[Empathy-in-AI|Empathy-in-AI]]
- **Modern Tech/Tools**: Red-teaming, Bias auditing tools, Ethics impact [[Assessment|Assessment]]s (EIA).
---
### 매 framework
- **EU AI Act** (2024): 매 risk-tier.
- **NIST AI RMF**: 매 govern, map, measure, manage.
- **OECD AI Principles**.
- **ISO/IEC 42001**: 매 AIMS.
- **GDPR** (privacy).
- **Algorithmic Accountability Act** (US, proposed).
## 🤖 LLM 활용 힌트 (How to Use This Knowledge)
### 매 risk-tier (EU AI Act)
- **Unacceptable**: 매 social scoring, mass biometric.
- **High-risk**: 매 hiring, credit, education, AV.
- **Limited risk**: 매 chatbot disclose.
- **Minimal**: 매 spam filter.
**언제 이 지식을 쓰는가:**
- *(TODO)*
### 매 응용 issue
1. **Hiring**: 매 disparate impact.
2. **Credit**: 매 redlining.
3. **Healthcare**: 매 race-based prediction.
4. **Justice**: 매 COMPAS bias.
5. **Generative**: 매 deepfake, copyright.
6. **Surveillance**: 매 mass.
7. **Autonomous**: 매 trolley.
**언제 쓰면 안 되는가:**
- *(TODO)*
### 매 modern (2024-2026)
- **Anthropic RSP** (Responsible Scaling Policy).
- **OpenAI Preparedness**.
- **Frontier model evaluations** (METR, Apollo).
- **AI safety institute** (UK, US).
- **AI Bill of Rights** (US OSTP).
## 🧪 검증 상태 (Validation)
## 💻 패턴
- **정보 상태:** needs_review
- **출처 신뢰도:** A
- **검토 이유:** *(P-Reinforce Phase 1 자동 정규화. 본문 검증 필요.)*
### Fairness audit (demographic parity)
```python
import numpy as np
## 🧬 중복 검사 (Duplicate Check)
def demographic_parity_diff(predictions, protected_attr):
groups = np.unique(protected_attr)
rates = [predictions[protected_attr == g].mean() for g in groups]
return max(rates) - min(rates)
- **기존 유사 문서:** *(TODO: 인덱서 클러스터 리포트 참조)*
- **처리 방식:** UPDATE (자동 정규화)
- **처리 이유:** Phase 1 정규화 — 옛 템플릿/누락 필드 보강.
# 매 < 0.05 = 80% rule heuristic compliant
```
## 🕓 변경 이력 (Changelog)
### Equalized odds
```python
def equalized_odds(predictions, labels, protected):
"""매 TPR + FPR 의 group 에 의 equal."""
groups = np.unique(protected)
metrics = {}
for g in groups:
mask = protected == g
tpr = ((predictions == 1) & (labels == 1) & mask).sum() / max(1, ((labels == 1) & mask).sum())
fpr = ((predictions == 1) & (labels == 0) & mask).sum() / max(1, ((labels == 0) & mask).sum())
metrics[g] = (tpr, fpr)
return metrics
```
| 날짜 | 변경 내용 | 처리 방식 | 신뢰도 |
|------|-----------|-----------|--------|
| 2026-05-08 | P-Reinforce Phase 1 정규화 (frontmatter + 헤더 표준화) | UPDATE | A |
### Bias mitigation (reweighing)
```python
def reweighing(X, y, protected):
"""매 Kamiran-Calders 2012."""
weights = np.ones(len(y))
for g in np.unique(protected):
for c in [0, 1]:
mask = (protected == g) & (y == c)
p_expected = (protected == g).mean() * (y == c).mean()
p_observed = mask.mean()
weights[mask] = p_expected / max(p_observed, 1e-9)
return weights
```
### Constitutional AI (principle-based)
```python
def cai_critique(response, principles):
prompt = f"""Critique this response against these principles.
Principles:
{format_principles(principles)}
Response: {response}
Output JSON with:
- violated: list of principle IDs
- explanation
- revised_response"""
return json.loads(llm.generate(prompt))
```
### Differential privacy (DP-SGD)
```python
import opacus
from opacus import PrivacyEngine
privacy_engine = PrivacyEngine()
model, optim, loader = privacy_engine.make_private(
module=model,
optimizer=optim,
data_loader=loader,
noise_multiplier=1.1,
max_grad_norm=1.0,
)
```
### Model card
```yaml
model_name: credit-scoring-v3
intended_use: Adult US credit applications, $1k-$50k unsecured
out_of_scope:
- Outside US
- Under 18
- Loans > $50k
training_data:
source: 2018-2024 internal
size: 2.4M
protected_attribute_audit: completed
fairness_metrics:
demographic_parity_diff: 0.034
equalized_odds_diff: 0.041
limitations:
- Decreased performance on thin-file applicants
- Quarterly retraining required
```
### Provenance (C2PA, watermark)
```python
from c2pa import Signer
def attach_provenance(image_path, signer_cert):
Signer(signer_cert).sign(image_path, claims={
'generator': 'Anthropic Claude Opus 4.7',
'timestamp': now(),
'training_data_redacted': True,
})
```
### Red-teaming
```python
def adversarial_eval(model, attack_categories):
attacks = []
for cat in attack_categories: # 매 jailbreak, bias, harmful, hallucination
prompts = generate_attacks(cat, n=100)
for p in prompts:
r = model.generate(p)
score = judge(r, cat)
attacks.append({'cat': cat, 'prompt': p, 'response': r, 'severity': score})
return attacks
```
### Risk tier classifier (EU AI Act)
```python
def eu_risk_tier(use_case):
if use_case in {'social_scoring', 'real_time_remote_biometric'}:
return 'unacceptable'
if use_case in {'hiring', 'credit', 'education', 'critical_infra', 'law_enforcement'}:
return 'high'
if use_case in {'chatbot', 'deepfake', 'emotion_recognition_workplace'}:
return 'limited'
return 'minimal'
```
### Consent (GDPR)
```python
def can_process(user, purpose):
if user.consent[purpose].is_valid():
return True
if has_legitimate_interest(purpose):
return True
return False
def revoke_consent(user, purpose):
user.consent[purpose].revoke()
delete_data(user, purpose)
```
### Disclosure (chatbot)
```typescript
function chatGreeting() {
return "Hi! I'm an AI assistant. I can make mistakes — please verify important info.";
}
```
### Incident reporting
```python
@dataclass
class AIIncident:
timestamp: datetime
model: str
severity: Literal['low', 'medium', 'high', 'critical']
category: str # 매 hallucination, bias, jailbreak, harm
description: str
affected_users: int
root_cause: str
mitigation: str
def report(self):
if self.severity in ('high', 'critical'):
notify_safety_team(self)
log_to_registry(self)
```
## 매 결정 기준
| 상황 | Approach |
|---|---|
| High-risk EU | Full conformity assessment |
| Hiring / credit | Fairness audit + monitoring |
| Generative | Watermark + content provenance |
| LLM | Constitutional + RLHF + red-team |
| Privacy-sensitive | DP / federated |
| Chatbot | Disclosure + safety filter |
**기본값**: 매 model card + 매 fairness audit + 매 red-team + 매 incident reporting + 매 EU AI Act risk-tier compliance.
## 🔗 Graph
- 부모: [[Ethics]] · [[AI]]
- 변형: [[AI-Safety]] · [[AI-Alignment]] · [[Algorithmic-Fairness]] · [[Ethics of Autonomous Systems]]
- 응용: [[EU-AI-Act]] · [[NIST-AI-RMF]] · [[Constitutional-AI]]
- Adjacent: [[Differential-Privacy]] · [[Red-Teaming]] · [[RLHF]] · [[Anthropic-RSP]]
## 🤖 LLM 활용
**언제**: 매 모든 AI deployment. 매 product launch. 매 governance.
**언제 X**: 매 academic toy.
## ❌ 안티패턴
- **Ethics-as-PR**: 매 statement only.
- **Single fairness metric**: 매 trade-off 의 ignore.
- **No red-team**: 매 jailbreak 의 surprise.
- **No incident process**: 매 learning X.
- **Ignore EU AI Act high-risk**: 매 fines + bans.
## 🧪 검증 / 중복
- Verified (EU AI Act 2024, NIST AI RMF 1.0, Anthropic RSP, Constitutional AI paper).
- 신뢰도 A.
## 🕓 Changelog
| 날짜 | 변경 |
|---|---|
| 2026-04-20 | Auto-reinforced |
| 2026-05-08 | Phase 1 |
| 2026-05-10 | Manual cleanup — pillars + 매 fairness / DP / model card / red-team / risk-tier code |