# ===================================================================
# MODULE 10, LESSON 4: ETHICAL LEADERSHIP AND RESPONSIBILITY
# ===================================================================

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime
import warnings
warnings.filterwarnings('ignore')

print("="*70)
print("ETHICAL LEADERSHIP AND RESPONSIBILITY IN FINANCIAL AI")
print("="*70)

# ----------------------------------------------------------------
# PART A: ETHICAL LEADERSHIP CHECKLIST
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART A: Ethical Leadership Checklist")
print("-"*60)

ethical_leadership = pd.DataFrame({
    'Principle': [
        'Set the tone for ethics',
        'Lead by example',
        'Empower others to act ethically',
        'Question assumptions',
        'Speak up about ethical concerns',
        'Educate others on ethical AI',
        'Engage diverse perspectives',
        'Integrate ESG into decision-making',
        'Champion fairness and transparency',
        'Hold yourself and others accountable'
    ],
    'Status': [
        'In Progress',
        'In Progress',
        'In Progress',
        'In Progress',
        'Achieved',
        'In Progress',
        'In Progress',
        'In Progress',
        'In Progress',
        'In Progress'
    ],
    'Next Action': [
        'Communicate ethical values in team meetings',
        'Document my decision-making process',
        'Delegate authority for ethical decisions',
        'Challenge one assumption this week',
        'Report an ethical concern to leadership',
        'Organise an ethical AI training session',
        'Include diverse voices in model reviews',
        'Evaluate models for ESG impact',
        'Advocate for model explainability',
        'Establish accountability mechanisms'
    ]
})

print("Ethical Leadership Checklist:")
print(ethical_leadership.to_string(index=False))

# ----------------------------------------------------------------
# PART B: RESPONSIBLE AI FRAMEWORK
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART B: Responsible AI Framework")
print("-"*60)

responsible_ai_framework = {
    "Governance": {
        "Description": "Organisational oversight for AI ethics.",
        "Activities": [
            "Establish an AI ethics committee.",
            "Develop AI ethics policies and guidelines.",
            "Provide ethics training for all AI practitioners."
        ],
        "Metrics": [
            "Number of ethical reviews conducted.",
            "Training completion rate.",
            "Policy compliance rate."
        ]
    },
    "Risk Management": {
        "Description": "Identify and mitigate ethical risks.",
        "Activities": [
            "Conduct AI impact assessments.",
            "Perform fairness and bias audits.",
            "Develop incident response procedures."
        ],
        "Metrics": [
            "Number of risks identified.",
            "Time to remediation.",
            "Number of incidents and resolutions."
        ]
    },
    "Technical": {
        "Description": "Build ethical AI systems.",
        "Activities": [
            "Implement fairness constraints.",
            "Use explainable AI (SHAP/LIME).",
            "Validate model robustness."
        ],
        "Metrics": [
            "Fairness metrics (disparate impact).",
            "Explanability coverage.",
            "Robustness test pass rate."
        ]
    },
    "Operational": {
        "Description": "Monitor and improve ethical performance.",
        "Activities": [
            "Monitor model drift and fairness drift.",
            "Conduct regular audits.",
            "Collect and act on feedback."
        ],
        "Metrics": [
            "Monitoring alert rate.",
            "Audit findings and resolutions.",
            "Stakeholder satisfaction."
        ]
    },
    "Cultural": {
        "Description": "Foster an ethical culture.",
        "Activities": [
            "Communicate ethical values.",
            "Recognise ethical behaviour.",
            "Encourage open dialogue."
        ],
        "Metrics": [
            "Employee engagement scores.",
            "Ethical behaviour recognition.",
            "Participation in ethics discussions."
        ]
    }
}

for layer, details in responsible_ai_framework.items():
    print(f"\n{layer}:")
    print(f"  {details['Description']}")
    print("  Activities:")
    for activity in details['Activities']:
        print(f"    • {activity}")
    print("  Metrics:")
    for metric in details['Metrics']:
        print(f"    • {metric}")

# ----------------------------------------------------------------
# PART C: ETHICAL DILEMMA DECISION LOG
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART C: Ethical Dilemma Decision Log")
print("-"*60)

dilemma_log = pd.DataFrame({
    'Date': ['2024-03-15', '2024-04-20', '2024-05-10'],
    'Dilemma': [
        'Using education level as a feature in credit model',
        'Not explaining model decisions to customers',
        'Collecting extra data without explicit consent'
    ],
    'Decision': [
        'Use a proxy (skill-based) and monitor fairness',
        'Implement SHAP explanations for all predictions',
        'Obtain explicit consent and limit data collection'
    ],
    'Rationale': [
        'Balances fairness and performance; protects against bias',
        'Transparency builds trust and meets regulatory expectations',
        'Respects customer privacy and complies with GDPR'
    ],
    'Outcome': [
        'Fairness ratio improved from 0.75 to 0.85',
        'Customer trust increased; regulatory risk reduced',
        'Compliant with data protection laws'
    ]
})

print("Ethical Dilemma Decision Log:")
print(dilemma_log.to_string(index=False))

# ----------------------------------------------------------------
# PART D: ESG INTEGRATION IN AI
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART D: ESG Integration in Financial AI")
print("-"*60)

esg_integration = {
    "Environmental": {
        "Opportunities": [
            "Climate risk models for loan portfolios.",
            "Green bond analytics.",
            "Carbon footprint tracking for investments."
        ],
        "Actions": [
            "Incorporate climate scenarios into stress testing.",
            "Develop ESG scoring models.",
            "Reduce computational energy use (Green AI)."
        ]
    },
    "Social": {
        "Opportunities": [
            "Fair lending for underserved communities.",
            "Financial inclusion models.",
            "Impact investing analytics."
        ],
        "Actions": [
            "Test models for disparate impact.",
            "Design models that serve all segments.",
            "Measure social impact of AI decisions."
        ]
    },
    "Governance": {
        "Opportunities": [
            "Transparent AI governance.",
            "Ethical AI policies.",
            "Stakeholder engagement."
        ],
        "Actions": [
            "Establish an AI ethics committee.",
            "Publish model cards and impact assessments.",
            "Engage with regulators and civil society."
        ]
    }
}

for pillar, details in esg_integration.items():
    print(f"\n{pillar}:")
    print("  Opportunities:")
    for opp in details['Opportunities']:
        print(f"    • {opp}")
    print("  Actions:")
    for action in details['Actions']:
        print(f"    • {action}")

# ----------------------------------------------------------------
# PART E: STAKEHOLDER ENGAGEMENT FOR ETHICAL AI
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART E: Stakeholder Engagement for Ethical AI")
print("-"*60)

stakeholder_engagement = {
    "Internal Stakeholders": {
        "Groups": [
            "Leadership", "Legal", "Compliance", "Risk", "Data Science", "IT"
        ],
        "Engagement": [
            "Regular ethics briefings for leadership.",
            "Legal and compliance review of AI policies.",
            "Risk assessments for AI projects.",
            "Ethics training for data science teams.",
            "Security and privacy review for AI systems."
        ]
    },
    "External Stakeholders": {
        "Groups": [
            "Customers", "Regulators", "Investors", "Community", "Academia"
        ],
        "Engagement": [
            "Transparent AI explanations for customers.",
            "Engage with regulators on AI policy.",
            "ESG reporting for investors.",
            "Community consultation on AI use cases.",
            "Collaboration with academic researchers."
        ]
    }
}

for group_type, details in stakeholder_engagement.items():
    print(f"\n{group_type}:")
    print("  Groups:")
    for g in details['Groups']:
        print(f"    • {g}")
    print("  Engagement:")
    for e in details['Engagement']:
        print(f"    • {e}")

# ----------------------------------------------------------------
# PART F: ETHICAL AI TRAINING PLAN
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART F: Ethical AI Training Plan")
print("-"*60)

training_plan = {
    "Target Audience": [
        "Data Scientists",
        "Data Engineers",
        "ML Engineers",
        "Product Managers",
        "Executives"
    ],
    "Topics": [
        "Introduction to Ethical AI",
        "Bias and Fairness",
        "Explainability and Transparency",
        "Privacy and Data Protection",
        "AI Governance and Compliance",
        "Case Studies in Financial AI Ethics"
    ],
    "Format": [
        "E-learning modules",
        "Workshops",
        "Case study discussions",
        "Guest lectures",
        "Ethics simulations"
    ],
    "Frequency": [
        "Annual refresher for all",
        "Quarterly for new hires",
        "Ongoing for leadership"
    ]
}

print("Ethical AI Training Plan:")
for category, items in training_plan.items():
    print(f"\n{category}:")
    for item in items:
        print(f"  • {item}")

# ----------------------------------------------------------------
# PART G: ETHICAL LEADERSHIP ACTION PLAN
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART G: Ethical Leadership Action Plan")
print("-"*60)

action_plan = {
    "Immediate (Next 30 Days)": [
        "Review and update AI ethics policies.",
        "Conduct a fairness audit of a current model.",
        "Organise an ethics discussion with your team.",
        "Identify and mitigate one ethical risk."
    ],
    "Short-Term (3-6 months)": [
        "Establish an AI ethics committee or working group.",
        "Implement explainability (SHAP/LIME) for all models.",
        "Develop a bias mitigation framework.",
        "Engage with regulators on AI ethics."
    ],
    "Medium-Term (1-2 years)": [
        "Integrate ESG into AI decision-making.",
        "Publish model cards for all models.",
        "Achieve certification in responsible AI.",
        "Contribute to industry standards on AI ethics."
    ],
    "Long-Term (3-5 years)": [
        "Become a recognised thought leader in ethical AI.",
        "Influence AI policy and regulation.",
        "Build an ethical AI culture across the organisation.",
        "Mentor the next generation of ethical AI leaders."
    ]
}

for period, actions in action_plan.items():
    print(f"\n{period}:")
    for action in actions:
        print(f"  • {action}")

# ----------------------------------------------------------------
# PART H: SUMMARY AND RECOMMENDATIONS
# ----------------------------------------------------------------

print("\n" + "="*70)
print("PART H: Summary and Recommendations")
print("="*70)

print("""
Ethical Leadership in Financial AI – Key Takeaways:

1. Ethical leadership is essential for building trust and sustainable AI.
2. Promote fairness, transparency, accountability, privacy, and robustness.
3. Establish a responsible AI framework: governance, risk, technical, operational, cultural.
4. Navigate ethical dilemmas with a structured decision-making framework.
5. Integrate ESG into AI to contribute to a sustainable future.
6. Engage stakeholders and build a culture of ethics.
7. Lead by example and empower others to act ethically.

Recommendations:
  - Start with self-reflection: what are your ethical values?
  - Communicate your commitment to ethical AI.
  - Invest in ethics training for your team.
  - Engage with diverse perspectives.
  - Be proactive in addressing ethical risks.
  - Celebrate ethical leadership and inspire others.
  - Continuously learn and improve your ethical practice.
""")

print("="*70)
print("END OF LESSON 4 – MODULE 10")
print("="*70)