Resources: Mechanistic interpretability

Exercises: Mechanistic interpretability

Resources: Rapidly testing your project

Exercises: Rapidly testing your project

Resources: Building in public

Exercises: Building in public

Next steps: Programs

Resources: Developing your project

Exercises: Developing your project

Resources: Technical governance approaches

Exercises: Technical governance approaches

Resources: AI and the years ahead

Exercises: AI and the years ahead

Resources: Contributing to AI safety

Exercises: Contributing to AI safety

Resources: What is AI alignment?

Exercises: What is AI alignment?

Resources: Further developing your project

Exercises: Further developing your project

Resources: Scalable oversight

Exercises: Scalable oversight

Resources: Robustness unlearning and control

Exercises: Robustness unlearning and control

Resources: Reinforcement learning from human (or AI) feedback

Exercises: Reinforcement learning from human (or AI) feedback

**This course has been replaced by our new [Technical AI Safety course](https://bluedot.org/courses/technical-ai-safety).**

How can you contribute?

AI safety is a young field with a lot of work to be done. Figuring out what to do can be difficult, as there aren't many default paths to follow.

This week encourages you to think about your career and goals, and to propose a project you could work on over the next four weeks to help you work towards those goals, informed by the knowledge you've gained throughout this course.

The main focus of this week are the two exercises, which you can find after the resources section. Having any plan helps resolve decision paralysis, and you will likely revisit and adapt this plan as you take action and learn more. The important thing, for now, is to get started.


Make a copy of this [Project Planning Template](https://docs.google.com/document/d/1-kGvm22AQTuVdethHD5u9tO_9xRkqKw2VztzG8qknA4/copy) and use the questions to help you start generating project ideas. Make the google doc [shareable](https://support.google.com/docs/answer/2494822?hl=en-GB&co=GENIE.Platform%3DDesktop#zippy=%2Cshare-a-file-publicly) and **paste the link** in the box below.

- **Before Session 8:** Complete section 1 to generate ideas for your project. We recommend spending at least 60 mins on this.
- **During Session 8:** You will get feedback on this document from your cohort, and begin crafting a plan to execute your project in section 2.

You are expected to get feedback from your cohort and facilitator. Based on [previous findings](https://bluedot.org/blog/is-proposal-feedback-helpful/), the BlueDot central team will **not** provide feedback on project documents by default.


Make your project awesome

During this course, you've learnt a lot about alignment, and you may now have a sense of the opportunities currently available. How might you factor this into your own career's big picture plan?
This exercise involves brainstorming different long-term paths you could pursue, similar to the suggestions by 80,000 Hours [here](https://80000hours.org/career-planning/process/longer-term-paths/). We encourage you to respond to the prompts below and list any uncertainties you have, and to submit your responses in the text box. The further resources may be particularly helpful.

- What are 3-5 roles you could have in 2-5 years, that seem important for AI safety?
    - You could start by brainstorming roles you think are needed to make AI safety go well, then evaluate what roles currently exist and what jobs are [available](https://jobs.80000hours.org/?refinementList%5Btags_area%5D%5B0%5D=AI%20safety%20%26%20policy) (the available jobs does not necessarily reflect the work that needs to be done!)
    - You could also consider roles adjacent or related to AI safety, that would help with developing skills and relationships.
- Which of these roles do you naturally feel most excited to test your fit for or work on?
- What [cheap tests](https://80000hours.org/career-guide/personal-fit/#do-cheap-tests-first) could help you actually test your fit for these different paths?

Optional other prompts:
- What strengths, skills, competencies, and other [career capital](https://80000hours.org/career-guide/career-capital/) do you already have? Set a [5-minute timer](https://www.google.com/search?q=5+minute+timer), and try to list at least 10 assets you have.
- What do you like and dislike about your current job? Have you had a past job you really enjoyed - if so, why do you think you enjoyed it?
- What goals did you write down in your icebreaker session? How can you now achieve them?
- Brainstorm other skills or areas of expertise that you could develop, that would enhance your ability to contribute. Try to focus on skills that are both [rare and valuable](https://www.scotthyoung.com/blog/2022/08/21/key-career-progress/). 
    - Another framing for this is to have [sets of skills](https://www.lesswrong.com/posts/XvN2QQpKTuEzgkZHY/being-the-pareto-best-in-the-world) where very few people have capabilities in both, but that can be leveraged together in high-value ways. 
    - We encourage you to consider options that you'd be excited to explore, or have a comparative advantage for.

To get feedback on your plan, we'd encourage sharing it with your cohort's Slack channel or just your facilitator.


AI Alignment

Exercises: Contributing to AI safety

Exercises