Joshua Clymer at the Centre for AI Safety will be giving an introduction to technical research in AI safety. The content will be simple enough for beginners at a high school level to understand.
Some topics covered include:
The main categories of AI safety research: making AI algorithms handle unexpected inputs, monitoring them for unwanted outcomes, setting proper goals for them, etc.
Examples of where AI safety research is needed today and where it might be needed in the future.
How AI safety research relates to other safety engineering and risk analysis disciplines.
Next steps for those interested in diving deeper into the field.
This talk is part of a series of talks on technological risks. Other talks in the series include:
An introduction to safety engineering by me. Link to details here.
Intro to AI Safety
Joshua Clymer at the Centre for AI Safety will be giving an introduction to technical research in AI safety. The content will be simple enough for beginners at a high school level to understand.
Some topics covered include:
The main categories of AI safety research: making AI algorithms handle unexpected inputs, monitoring them for unwanted outcomes, setting proper goals for them, etc.
Examples of where AI safety research is needed today and where it might be needed in the future.
How AI safety research relates to other safety engineering and risk analysis disciplines.
Next steps for those interested in diving deeper into the field.
This talk is part of a series of talks on technological risks. Other talks in the series include:
An introduction to safety engineering by me. Link to details here.
An introduction to cyberbiosecurity risks by Kathryn Millet at Biosecure Ltd. Link to details here.
The event will be hosted via Zoom. If you’re interested in joining, message @Madhav Malhotra for a link.