NIST has published a report, Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations. This is intended to help developers protect Chatbots and Self-Driving Cars from Digital Threats by understanding the types of attacks to expect and approaches to mitigate them.
The report covers two broad types of AI: predictive AI and generative AI and identifies four major types of attacks on AI systems:
- Evasion attacks: These occur after an AI system is deployed, where a user attempts to alter an input to change how the system responds to it.
- Poisoning attacks: These occur in the training phase through the introduction of corrupted data.
- Privacy attacks: These occur during deployment and they are attempts to learn sensitive information about the AI or the data it was trained on with the goal of misusing it.
- Abuse attacks: These involve inputting false information into a source from which an AI learns.
Defensive measures include, but are not limited to:
- Augmenting the training data with adversarial examples
- Monitoring standard performance metrics for degradation in classifier metrics
- Using data sanitization techniques
Troy Batterberry, CEO and Founder, EchoMark had this comment:
“NIST’s adversarial ML report is a helpful tool for developers to better understand AI attacks. The taxonomy of attacks and suggested defenses underscores that there’s no one-size-fits-all solution against threats; however, understanding of how adversaries operate, and preparedness are critical keys to mitigating risk.
“As a company who uses leverages AI and LLMs as part of our business, we understand and encourage this commitment to secure AI development, ensuring robust and trustworthy systems. Understanding and preparing for AI attacks is not just a technical issue but a strategic imperative necessary to maintain trust and integrity in increasingly AI-driven business solutions.”
Guidance like this is always helpful. But it’s only helpful if this guidance is followed. Thus I hope the target audience of this report are paying attention and follow this guidance as that will make us all safer.
Like this:
Like Loading...
Related
This entry was posted on January 6, 2024 at 8:25 am and is filed under Commentary with tags NIST. You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.
NIST Publishes Adversarial Machine Learning Playbook For Developers
NIST has published a report, Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations. This is intended to help developers protect Chatbots and Self-Driving Cars from Digital Threats by understanding the types of attacks to expect and approaches to mitigate them.
The report covers two broad types of AI: predictive AI and generative AI and identifies four major types of attacks on AI systems:
Defensive measures include, but are not limited to:
Troy Batterberry, CEO and Founder, EchoMark had this comment:
“NIST’s adversarial ML report is a helpful tool for developers to better understand AI attacks. The taxonomy of attacks and suggested defenses underscores that there’s no one-size-fits-all solution against threats; however, understanding of how adversaries operate, and preparedness are critical keys to mitigating risk.
“As a company who uses leverages AI and LLMs as part of our business, we understand and encourage this commitment to secure AI development, ensuring robust and trustworthy systems. Understanding and preparing for AI attacks is not just a technical issue but a strategic imperative necessary to maintain trust and integrity in increasingly AI-driven business solutions.”
Guidance like this is always helpful. But it’s only helpful if this guidance is followed. Thus I hope the target audience of this report are paying attention and follow this guidance as that will make us all safer.
Share this:
Like this:
Related
This entry was posted on January 6, 2024 at 8:25 am and is filed under Commentary with tags NIST. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.