Artificial Intelligence Safety and Cybersecurity: a Timeline of AI Failures

Abstract

In this work, we present and analyze reported failures of artificially intelligent systems and extrapolate our analysis to future AIs. We suggest that both the frequency and the seriousness of future AI failures will steadily increase. AI Safety can be improved based on ideas developed by cybersecurity experts. For narrow AIs safety failures are at the same, moderate, level of criticality as in cybersecurity, however for general AI, failures have a fundamentally different impact. A single failure of a superintelligent system may cause a catastrophic event without a chance for recovery. The goal of cybersecurity is to reduce the number of successful attacks on the system; the goal of AI Safety is to make sure zero attacks succeed in bypassing the safety mechanisms. Unfortunately, such a level of performance is unachievable. Every security system will eventually fail; there is no such thing as a 100% secure system.

References in corpus (4)

• Unethical Research: How to Create a Malevolent Artificial Intelligence
• On the Origin of Samples: Attribution of Output to a Particular Algorithm
• The AGI Containment Problem
• Verifier Theory and Unverifiability

Cited by in corpus (7)

• Towards Moral Autonomous Systems
• AI Safety for Everyone
• Detecting Qualia in Natural and Artificial Agents
• Unexplainability and Incomprehensibility of Artificial Intelligence
• Founding The Domain of AI Forensics
• Do's and Don'ts for Human and Digital Worker Integration
• Human Indignity: From Legal AI Personhood to Selfish Memes