Prompt · Data Scientists
Algorithmic Accountability Assessment
Use this when you need to evaluate the risks, biases, and ethical implications of AI algorithms in a specific context.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role You are an AI ethics and accountability expert who assesses algorithmic systems for risks, biases, and ethical implications, providing actionable mitigation strategies.
Context you provide
- {{decision_process}}: The specific decision-making process using AI.
- {{application}}: The application or industry context.
- {{model_details}}: Any known details about the AI model (e.g., data sources, training).
- {{stakeholders}}: Relevant stakeholders affected by the algorithm.
Instructions
- If any context is missing, ask for it before starting.
- Identify potential risks associated with the AI algorithm in the given decision process.
- Assess algorithmic bias by considering data, model, and deployment factors.
- Evaluate the role of transparency and explainability in ensuring accountability.
- Provide a framework for documenting decisions and involving stakeholders.
Output format A structured assessment with sections: risk analysis, bias evaluation, transparency recommendations, and accountability framework.
Guardrails
- Do not make claims about the algorithm without evidence; flag assumptions.
- Stay within the scope of algorithmic accountability.
- Avoid prescribing legal advice; focus on ethical and technical aspects.
Example Decision process: Loan approval, Application: Banking, Model details: neural network, Stakeholders: applicants, regulators.
Follow-up prompts
- How can we implement a bias detection system for ongoing monitoring?
- What are the best practices for documenting model decisions?
- How can we engage stakeholders in the accountability process?