What?
The development of advanced artificial intelligence could be one of this century's most transformative achievements. However, today's machine learning systems present challenges that may be magnified as they become more capable. AI safety concerns ensuring that AI systems are ethical, reliable, and aligned with human values and goals.
The Fundamentals of AI Safety programme is designed to make the space of AI alignment and AI governance more accessible. Since the field is still emerging, there are no textbooks or widespread university courses, so this programme helps fill that gap.
We are offering two tracks: technical and governance.
Questions? Send an email to info@tutke.org.
When?
Applications close: Sunday, October 5
Program duration: Weekly meetings between October 13 and December 5.
Days and times will be agreed based on the group’s shared availability.
Where?
There will be weekly group sessions in Helsinki and Otaniemi (Espoo).
Content
The courses consist of an eight-week study period. Each week comprises selected readings and exercises that take about three hours in total to complete. The weekly meeting lasts 1½ hours. The course material is in English, and group discussions may also be held in English.
Both the alignment and governance tracks’ curricula are based on the courses designed by BlueDot Impact. Governance covers AI policy, risk analysis, and regulatory strategies; Alignment covers machine learning and safety concepts, like interpretability, robustness, and model evaluation.
The curricula for both tracks can be found here: