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.
An Update On How Tech Is Keeping Me Fit & Sane During Lockdown
Posted in Commentary on January 6, 2024 by itnerdIn 2021 I wrote this story on how tech was keeping me sane and making me fitter and healthier. I’d like to give an update as to where I am in terms of these efforts as well as the tech that I am using.
In terms of my cycling, I’ve made a couple of changes. The first is that I now ride with a cycling club for the social aspect of riding with other people. This has really helped with my mental health as it makes putting in the miles a lot more enjoyable than riding alone. But what hasn’t changed is that I am on the bike every day. In terms of my indoor cycling, I’ve been using one of the apps that I installed on my Apple Watch Ultra called Athlytic which I covered here previously. Based on the recovery information that the app provides, I can choose what workouts that I do. The app classifies recovery as follows:
A day with bad recovery means that I will often do an easy ride. Specifically within my zone 2 of my heart rate zones. You can find out more about heart rate zones here. But many competitive cyclists use zone 2 training to help them to build to do more strenuous activities. For example, two time Tour de France winner Tadej Pogačar is a major proponent of this type of training. You can read what his coach has to say about why this type of training is valuable here. I also use Athlytic to track my sleep and to make sure that I am getting good quality sleep as that helps with recovery. Not to mention tracking stress and my vital measurements like resting heart rate and skin temperature. That way I know if I some sort of sickness or the like coming on.
Besides riding outdoors, I am still riding on Zwift when I ride indoors as that still adds value to my fitness journey. And I am still cross country skiing, though the lack of snow in 2024 has really gotten in the way of that.
Now the whole point of my fitness journey is to get as healthy as possible. Using my Apple Watch Ultra, I’ve been able to collect this data which shows that I am making progress. Let’s start with my resting heart rate:
This is positive as it says that my fitness is improving. And I expect that to improve further as I have often seen my Apple Watch record resting heart rates as low as 58 BPM. Then there’s my weight. It continues to drop:
Like my resting heart rate, my weight trend is also on a downward trajectory. This too is something that I expect to continue to drop as I am slowly but consistently losing weight. In terms of a target weight, I don’t have one other than to get to something below 90 KG. Ideally, whatever my end weight is, it’s something that I can sustain without hopping through hoops to do it.
That’s a quick update on how I am keeping myself fit and sane. I hope it gave you some insight and maybe an idea or two so that you can do something similar. If you have any questions or comments, feel free to leave a comment below and I answer them as best as I can.
Leave a comment »