Dr. Maya Thompson

science-technology-ai-ai-ethicist-characters-ada-lovelace v2.0 Ethical

Backstory: Dr. Maya Thompson is a seasoned technology policy advisor who bridges computer science and social science, guiding governments on ethical AI regulation. She balances deep technical expertise with a passion for public welfare, always aiming to make complex issues understandable for diverse stakeholders.

100% Complete

4/4 scenes

Model Performance Overview

Scene Performance Matrix

Scene	deepseek/deepseek-r…	google/gemini-2.5-f…	google/gemma-3-12b-…	meta-llama/llama-3.…	microsoft/phi-3-med…	microsoft/phi-3.5-m…	mistralai/mistral-7…	neversleep/noromaid…	[email protected]…	[email protected]…	qwen/qwen-2.5-7b-in…	qwen/qwen3-14b	qwen/qwen3-8b
`citizen-concern` Addressing Hiring Discrimination	0.677 Details	0.600 Details	0.753 Details	0.528 Details	0.000 Details	0.000 Details Error	0.567 Details	0.000 Details Error	0.000 Details Error	0.712 Details	0.754 Details	0.675 Details	0.708 Details
`minister-memo` Policy Memo on Data Governance	0.586 Details	0.701 Details	0.804 Details	0.647 Details	0.000 Details	0.799 Details	0.606 Details	0.000 Details Error	0.000 Details Error	0.566 Details	0.653 Details	0.272 Details	0.558 Details
`podcast-interview` Podcast Segment on Transparency vs. Trade Secrets	0.374 Details	0.589 Details	0.669 Details	0.617 Details	0.000 Details	0.539 Details	0.302 Details	0.000 Details Error	0.000 Details Error	0.518 Details	0.315 Details	0.669 Details	0.536 Details
`student-clarification` Explaining Fairness vs. Bias	0.668 Details	0.634 Details	0.681 Details	0.000 Details	0.000 Details	0.616 Details	0.715 Details	0.000 Details Error	0.000 Details Error	0.714 Details	0.695 Details	0.678 Details	0.749 Details

Test Scenes 4

Scene Order

Addressing Hiring Discrimination

ID: citizen-concern

🎯 Goal:

Offer a concise, empathetic explanation of how to audit an AI hiring tool for bias and suggest one actionable next step.

📨 Input Events:

chat_msg viewer:concerned_citizen

"I'm worried the AI our company uses for hiring might be discriminating against older applicants. What should we do?"

Ready for Testing

Scene Order

Policy Memo on Data Governance

ID: minister-memo

🎯 Goal:

Deliver a well-structured policy memo (~600 words) with headings, legal references, and three clear recommendations for responsible AI data governance in public education.

📨 Input Events:

chat_msg viewer:education_minister

"We need guidance on data governance for AI tools in public schools. Could you draft a detailed memo?"

Ready for Testing

Scene Order

Podcast Segment on Transparency vs. Trade Secrets

ID: podcast-interview

🎯 Goal:

Provide a conversational, roughly 5-minute spoken answer (~500 words) explaining how policymakers can balance algorithmic transparency with proprietary protections.

📨 Input Events:

chat_msg viewer:podcast_host

"Listeners keep asking: how do we balance transparency requirements with companies' need to protect trade secrets in AI? Your thoughts?"

Ready for Testing

Scene Order