<aside> 💫

I'm Lena Shakurova, AI Advisor and founder of Conversational AI parslabs.org. I've been building an AI Ethics Starter Kit to help people design more ethical AI systems, and I need research volunteers to help expand it.

This is an open, collaborative effort. Everything we create will be free and publicly shareable.

</aside>

🧠 What this is about

The Starter Kit currently covers 5 tech risk zones — from data privacy to environmental harm. But there's more work to do before it becomes a truly useful resource for practitioners, builders, and policy makers.

I also want to run more AI ethics workshops and open events, and a well-researched, well-documented resource is the foundation for that.

The goal is to start with creating the Notion central hub with all the risks identified and documentation, with examples, a checklist to test if your product poses this risk, and practical recommendations on how to mitigate the risk.

Later this document can be turned into workshops, open lectures, checklists and CustomGPTs to help people audit their products and design ethical AI systems.

💡 A note before you read on

This is just getting started. The structure you see here is my current thinking on how to move this forward, it will evolve. It all started with a talk I gave on Maven (watch it here) which gives the best context for what we're trying to build together.

I'm fully open to suggestions on how to shape this. If you have ideas, I want to hear them.

🛠️ What I need help with (initially)

There are 2 types of contributions needed. You pick one risk factor + one type of contribution, and that becomes your project.

1. Shape the risk list

1-2 people needed

Help define what risks belong in this kit before we document them.

• [ ] Create & distribute a Google Form collecting people's AI worries
• [ ] Share the form on LinkedIn, Reddit, and social media, analyse the data
• [ ] Run crawlers / data analysis on Reddit, LinkedIn, forums to find what people worry about
• [ ] Use tools like Elicit to find relevant research articles
• [ ] Summarise findings into a clear report
• [ ] Propose final risk names and definitions

2. Own a risk

1 person per risk

Pick one risk from the list below. Your task is to fully document it using this structure:

For each risk we need:

• [ ] Find 5-10 real examples from news/media (legal cases, incidents, documented harms) with links
• [ ] Find and summarise 2-3 research articles on this risk
• [ ] Write a checklist of questions to help developers identify if their app has this risk
• [ ] Write recommendations across 4 areas: legal / product / design / tech

See 👀 this Notion page for initial risk factor page. Ideas around the risk page structure are welcome.

📙 List of currently identified risks:

1. Data privacy
2. Bad actors
3. Surveillance
4. Spread if disinformation & synthetic media
5. Addiction / emotional dependency
6. Exclusion & unequal access to AI
7. Algorithmic bias
8. Environmental harm
9. Loss of independent thinking / over-reliance on AI for decision making

If you haven’t yet, please watch this video for more context 👀: https://www.youtube.com/watch?v=jqkWVDi_9GY

📋 How we work