A short video, a carousel, an illustrated explainer or a podcast segment: make one piece of honest, accurate AI safety content for an audience you understand.
BlueDot has argued that AI safety needs more people making content for general audiences, who are unlikely to read research papers.
Output: One published piece of content.
Your course-mates, a hobby community, people who speak your language. Content works best when you are part of the audience.
Choose one concept or one finding from a project on this site. One idea per piece.
Read the original source, not just headlines. Note exactly what is and is not known.
Script or storyboard it first, then produce it.
Share it, then note what people asked about or misunderstood. That’s your next piece.
All steps done
Share what you made, then try another project. The PDF version has room for notes.
The International AI Safety Report is the shared scientific picture of AI risk, written for governments. Turn one chapter into a short, local-audience summary.
Write a short, jargon-free explainer of one AI safety idea (reward hacking, sycophancy, compute governance, DNA screening) for someone outside tech.
Explain how liability and insurance could shape AI companies’ behaviour, using one realistic harm scenario to make it concrete.
