The malicious use of artificial intelligence: Forecasting, prevention, and mitigation
Злонамеренное использование искусственного интеллекта: прогнозирование, предотвращение и минимизация последствий
2018-02-20
SCID: 54.1/wa352deq
Discuss with AI
AI security threatsmalicious use of artificial intelligencethreat forecastingthreat mitigationthreat prevention
Figures from the paper
Abstract (AI)
This report surveys the landscape of potential security threats from malicious uses of AI, and proposes ways to better forecast, prevent, and mitigate these threats. After analyzing the ways in which AI may influence the threat landscape in the digital, physical, and political domains, we make four high-level recommendations for AI researchers and other stakeholders. We also suggest several promising areas for further research that could expand the portfolio of defenses, or make attacks less effective or harder to execute. Finally, we discuss, but do not conclusively resolve, the long-term equilibrium of attackers and defenders.
Key Findings
1
It proposes high-level recommendations for AI researchers and other stakeholders to improve forecasting, prevention, and mitigation of AI-enabled threats.
2
The long-term equilibrium between AI-enabled attackers and defenders remains unresolved, according to the report.
3
The report identifies promising research directions for expanding defensive capabilities and making malicious attacks less effective or harder to execute.
4
The report surveys potential security threats arising from malicious uses of artificial intelligence across digital, physical, and political domains.
Research Object
Malicious uses of artificial intelligence
Research Subject
security threats and the forecasting, prevention, and mitigation of those threats
Publication Details
Publication Date
2018-02-20
Journal
Publisher
ISSN
Cited by
525
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest