Introduction to AI Safety

Introduction to AI Safety

Instructor: Dan Hendrycks

3 modules

Gain insight into a topic and learn the fundamentals.

Beginner level

No prior experience required

4 hours to complete

Flexible schedule

Learn at your own pace

3 modules

Gain insight into a topic and learn the fundamentals.

Beginner level

No prior experience required

4 hours to complete

Flexible schedule

Learn at your own pace

What you'll learn

The foundations of modern AI systems, and where to expect continued development
About major risks from AI, including malicious use, competitive pressures, and loss of control
Governance strategies for mitigating AI risks across different scales

Details to know

See how employees at top companies are mastering in-demand skills

Learn more about Coursera for Business

logos of Petrobras, TATA, Danone, Capgemini, P&G and L'Oreal

There are 3 modules in this course

This course is an accessible introduction to core concepts in AI safety. We'll cover the drivers of modern AI development, major risks from AI like malicious use, competitive pressures, and loss of control, and governance strategies across different scales for preventing these risks.

This module introduces modern Artificial Intelligence: what AI can do today, how AI got here, drivers and constraints on AI development, and Artificial General Intelligence (AGI).

What's included

8 videos1 reading1 assignment1 discussion prompt

8 videosTotal 39 minutes

Welcome to 'Introduction to AI Safety'2 minutes
1-1: What Can AI Do Today? 6 minutes
1-2: How Did AI Get Here?11 minutes
1-3: Drivers of AI Development7 minutes
1-4: Constraints on AI Development4 minutes
1-5: What is Artificial General Intelligence (AGI)7 minutes
1-6: Timelines to AGI1 minute
1-7: Module Recap2 minutes

1 readingTotal 5 minutes

(Optional) Module 1 Additional Readings5 minutes

1 assignmentTotal 10 minutes

Module 1 Quiz10 minutes

1 discussion promptTotal 30 minutes

Module 1 Discussion Forum30 minutes

This module looks at why AI might be dangerous. We introduce the three main categories of risks from AI: malicious use, competitive pressures, and loss of control.

What's included

7 videos1 reading1 assignment1 discussion prompt

7 videosTotal 32 minutes

2-1 Malicious Use of AI12 minutes
2-2 Competitive Pressures5 minutes
2-3 Loss of Control - Erosion of Control3 minutes
2-4 Loss of Control - Rogue AI4 minutes
2-5 Loss of Control - Intelligence Explosion4 minutes
2-6 Loss of Control - Power Concentration1 minute
2-7 Module Recap2 minutes

1 readingTotal 5 minutes

(Optional) Module 2 Additional Readings5 minutes

1 assignmentTotal 10 minutes

Module 2 Quiz10 minutes

1 discussion promptTotal 30 minutes

Module 2 Discussion Forum30 minutes

This module looks at governance strategies for mitigating AI risks. We’ll look at governance interventions from the AI system level up to the international level, and evaluate their respective benefits and challenges.