The 5 Ws of AIDA
This is the second in a four part series on the Artificial Intelligence and Data Act (AIDA). The first post has some vital history on human rights and previous efforts to legislate tech in canada, this second post is not about analysis but rather a summary of what the bill is. The third and fourth will be analysis and ways to engage on it.
A small group of read the 18 pages of AIDA, and the preamble to it, over the weekend. We have a public copy for mark-up available here. We have not marked it up much, but in the spirit of trying to make this bill something more of us feel confident engaging in, we’ve got it as a starting point and the reading is done.
As someone that does not have a background in law, reading a bill like this is a bit daunting and the language is not accessible in any kind of a general way. So what follows here are some very basic things we learned about the bill.
And as a necessary comment, it is not ideal to be hiving this assessment of AIDA off from Bill C-27 as a whole, as there are important connections to be drawn between them, but there is also benefit to taking this smaller bite as AIDA has been packaged as a stand-alone within C-27 by ISED.
AIDA was created by, and draws often on the authority of, Innovation, Science, and Economic Development Canada. Also known as ISED.
ISED, in their own words: “Innovation, Science and Economic Development Canada (ISED) works with Canadians in all areas of the economy and in all parts of the country to improve conditions for investment, enhance Canada’s innovation performance, increase Canada’s share of global trade and build a fair, efficient and competitive marketplace. We are the federal institution that leads the Innovation, Science and Economic Development portfolio.”
AIDA has two parts. the bulk of it, under part one, is listed as “Regulation of Artificial Intelligence Systems in the Private Sector”. part one makes up over 90% of the bill, so at a very high level, this bill can be considered as having a focus of regulating the private sector. the remainder, which falls under part two, is listed as: “General Offences Related to Artificial Intelligence Systems”.
There are important nuances in how this relates to public sector use of AI, and how the government is managing that through other legislation and process. But for now, it’s roughly accurate to say that the major intent of AIDA is focused on trying to influence private sector design, development, and use of AI.
Here is the stated purpose of the bill, which is two-fold as well, as written in AIDA:
(a) to regulate international and interprovincial trade and commerce in artificial intelligence systems by establishing common requirements, applicable across Canada, for the design, development and use of those systems; and
(b) to prohibit certain conduct in relation to artificial intelligence systems that may result in serious harm to individuals or harm to their interests.”
When and Where
Soon. This bill is currently a proposed bill. That means it is in draft. The next step for the bill is for it to go to its second reading at the House of Commons. The next step for us to follow, from a legislative process perspective, is for when it goes to go to a federal committee. From history, the bill was supposed to go to INDU —the Standing Committee on Industry and Technology”. When I asked INDU about how to follow the timing of this bill and when it was coming to committee so we could engage with it there, I received this reply, on October 6, 2022:
“The committee has not yet had Bill C-27 referred and therefore has no study on this subject. The Bill has yet to be debated at Second Reading in the House. At this point, I would ask that you please circle back once the Bill has been referred to committee before requesting to appear. It might even be possible for this bill to be referred to a different committee. We won’t know which committee will be selected until the motion is moved for the Second Reading of the Bill.”
The only thing that those of us keep hearing is that it is headed to committee soon. So as of this date, we do now know exactly when, and we do not know which committee it will go to for consideration.
It is unclear why AIDA was created. It is particularly unclear as to how it made it out ISED’s door, given how rough the bill is, as it stands now. It would be helpful for ISED to disclose more about the process that went into the creation of this bill. Who wrote it and who was consulted? When ISED did a public consultation process for the Digital Charter back in 2018, it did not make much reference to artificial intelligence, nor to AIDA. It is safe to say that there was not fulsome general public consultation on this matter.
In the purpose of the bill, there is reference to the need to regulate international and interprovincial trade for artificial intelligence systems. Other countries have been creating legislation around AI, so this sense of needing to do what others are doing may be part of it. The intent here is not to guess or to analyze, but to simply state that beyond the “what” provided by ISED, the more common or public answer to “why” is not clear to me.
Parts three and four of this series will get into analysis/what to about this bill, including a look at some of the terms and constructs and frames that appear to be problematic with how the bill is currently drafted. These include terms such as “high-impact” systems, a focus on the individual over the collective, questions of governance, questions of access to justice, questions of harm, and more.