Evidence of meeting #53 for Access to Information, Privacy and Ethics in the 44th Parliament, 1st Session. (The original version is on Parliament’s site, as are the minutes.) The winning word was aws.
A recording is available from Parliament.
4:50 p.m.
Bloc
René Villemure Bloc Trois-Rivières, QC
Thank you, Mr. Chair.
Thank you for being here today, Ms. Foster.
I use technology, but I'm not an expert like you.
Who are AWS's competitors?
4:50 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
Within the cloud framework agreement there are eight other companies that participate as organizations offering cloud services to the Government of Canada. I'm not sure if I can reel all of them off by heart, but they include Google, Microsoft, ThinkOn, ServiceNow, Oracle, IBM and companies like that.
4:50 p.m.
Bloc
4:50 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
Yes.
4:50 p.m.
Bloc
René Villemure Bloc Trois-Rivières, QC
All right.
You said you had a certain number of tools to secure data.
I would like you to be more specific. Tell us about your security controls, your tools. What exactly do you do? It's a bit too general for me, right now.
4:50 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
There are a couple of different ways of looking at or considering that. One is looking at the broader cloud environment, in which we secure the cloud itself. There's also the physical security of our data centres. We manage the physical security of those sites and ensure that they're not at risk of natural disasters, that they have access to reliable energy, that they are not on a flood plain, and things like that.
There's the physical security of the data centre. There's the virtual security of the cloud. These are creating a more secure environment.
In terms of security tools, there are quite a lot of different security tools. A couple of examples of those would be...for example, we have some artificial intelligence tools that are able to monitor for what might be unusual access to a customer's data or a customer's environment. We have systems that would alert us, and then we can alert a customer. Sometimes it turns out to be nothing and sometimes it turns out to be something, but we have the ability to use artificial intelligence to monitor and alert to those things.
We also provide encryption tools. That's probably one of the most important security tools we offer. Customers always have the choice to encrypt their data. Once it's encrypted, they also hold the encryption keys to their data.
4:50 p.m.
Bloc
René Villemure Bloc Trois-Rivières, QC
When AWS worked on the ArriveCAN project—I'm not talking about the app—was it like any other project, or was it unique?
4:50 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
That's a good question.
There are so many unique use cases for cloud services, and none of them are quite the same. I would say that the uniqueness of this would include a couple of features. One was the high security bar that was required for this project. The other is the speed with which it needed to be executed, with that high bar of security. Being able to incorporate a number of different functionalities so quickly made it unique.
The other thing that was going on at the time was...of course, we all remember that public health guidance and policies on what you needed in order to travel were changing pretty frequently. The ability to operationalize those changes into the technology, also at speed, also made this project somewhat unique.
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
No. It wasn't huge. There were a number of people who were brought in very quickly to support it, but it wasn't a huge team, no.
4:55 p.m.
Bloc
René Villemure Bloc Trois-Rivières, QC
An article online mentioned a partner by the name of Fortinet. Can you tell me about Fortinet?
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
Do you mean in terms of involvement on ArriveCAN?
4:55 p.m.
Bloc
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
I don't have any knowledge of direct work with Fortinet. I'm sorry.
4:55 p.m.
Bloc
René Villemure Bloc Trois-Rivières, QC
It was mentioned in an article on Amazon's site.
Fortinet wasn't a company you worked with regularly, then?
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
I'm not sure about its being directly related to ArriveCAN, but we work with a number of different partners who help customers implement services or manage different aspects of projects they might have. It could include companies like that. It includes companies such as Accenture. Deloitte would be another example.
We work with a number of different companies that help our customers adopt the technology.
4:55 p.m.
Bloc
René Villemure Bloc Trois-Rivières, QC
Do people have the same confidentiality or privacy requirements?
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
I think it would vary, depending on the project and what the requirements of that specific project were.
4:55 p.m.
Bloc
René Villemure Bloc Trois-Rivières, QC
You said at the beginning of your opening remarks that the servers were located in Canada. Are they all in Canada, or just some of them?
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
I'm sorry. Is what strictly in Canada?
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
We have data centre infrastructure all over the world. We enabled the customer—in this case, ArriveCAN—to do the whole project within Canada. In terms of what was ultimately decided, that would be up to the Government of Canada to determine where it put its data.
The customer has full control over how they operationalize in the cloud.
4:55 p.m.
Bloc
4:55 p.m.
Director, Global Artificial Intelligence and Machine Learning, Canada Public Policy, Amazon Web Services, Inc.
It wouldn't be a mystery as to what the Government of Canada chose. It would be up to the Government of Canada to give you that information, but it is enabled to exist completely in Canada.