Published Jan 11, 2022

Eliminate AI failures

AI expert Yaron Singer delves into the indispensable role of human judgment in AI to prevent failures, addresses vulnerabilities like adversarial attacks and distributional drift, and proposes AI firewall solutions to secure models by monitoring data to ensure fair and robust predictions.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

When AI goes wrong
Answers 383 questions
Success (and failure) in prompting
Answers 383 questions
Achieving provably beneficial, human-compatible AI
Answers 383 questions
AI's impact on developers
Answers 383 questions
Creating tested, reliable AI applications
Answers 383 questions
Fighting bias in AI (and in hiring)
Answers 383 questions
AI adoption in the enterprise
Answers 383 questions
Explainable AI that is accessible for all humans
Answers 383 questions
So you have an AI model, now what?
Answers 383 questions
Generative models: exploration to deployment
Answers 383 questions
Balancing human intelligence with AI
Answers 383 questions
Testing ML systems
Answers 383 questions
AIs that look human and create portraits of humans
Answers 383 questions
The last mile of AI app development
Answers 383 questions
Practical AI Ethics
Answers 383 questions

Dexa/Practical AI

Eliminate AI failures

Topics covered

Popular Clips

Streamlined Model Integration

AI Model Failures

AI Decision Making

Future of AI

Challenges in Machine Learning

Journey to Robust Intelligence

AI Insights

Understanding AI Failures

Mitigating AI Failures

AI Risk Perception

Testing for Bias

AI Automation Insights

AI Firewall Testing

Decoupling for Robustness

Episode Highlights

Human-AI Collaboration

AI Model Vulnerabilities

AI Firewall Solutions

Related Episodes

When AI goes wrong

Success (and failure) in prompting

Achieving provably beneficial, human-compatible AI

AI's impact on developers

Creating tested, reliable AI applications

Fighting bias in AI (and in hiring)

AI adoption in the enterprise

Explainable AI that is accessible for all humans

So you have an AI model, now what?

Generative models: exploration to deployment

Balancing human intelligence with AI

Testing ML systems

AIs that look human and create portraits of humans

The last mile of AI app development

Practical AI Ethics

Eliminate AI failures

Topics covered

Popular Clips

Episode Highlights

Human-AI CollaborationYaron Singer discusses the critical role of human judgment in AI systems and the inherent risks of increased automation. He emphasizes the need for a balanced approach to ensure AI supports rather than replaces human decision-making.

Human-AI Collaboration

AI Model VulnerabilitiesYaron Singer explores the vulnerabilities of AI models, focusing on adversarial attacks and distributional drift. He discusses strategies for risk abstraction to enhance AI security and prevent failures.

AI Model Vulnerabilities

AI Firewall SolutionsYaron Singer discusses the implementation of AI firewalls to safeguard models from errors and biases. By decoupling model building from security, these firewalls monitor data, ensuring robust and fair AI predictions.

AI Firewall Solutions

Related Episodes