Hello advisors. Sorry for dropping this for a couple of weeks. I just looked at it again and still think it would a worthwhile exercise. There is a good set of questions to consider. We already have a couple of responses on record for some of the questions they are asking -- from Garth and from Evan. I don't think there is any harm in crafting a response which we could send into the process if we can get a few people on a zoom call. Whether such a response moves any dials in the process that is underway doesn't matter that much to me. All the responses will be visible on their website, so our comments would remain intact in that way. We can also put the responses on our website. And Evan's suggestion of sending something directly to ISED could also happen.
Let's try to find a time to have a one hour call and see what we can come up with.
Marita
Hi Marita,
Thanks for sending this.
Given the tone of the original letter, I'd be extremely hesitant to participate further with this org unless there's some confidence that "the People's Consultation" is really approaching the issue with an open perspective. I've seen previous examples of this tactic in which *any* submission to their effort just adds submittors to the list of signatories, asserting endorsement of whatever their next attempt produces; indeed it may be no different from the original version that some TC members found problematic.
Frankly, one could use their template to simply craft a TC response directly to ISED, and bypass the gatekeeper altogether. I don't have the cycles to do this single-handedly but would be fine participating in a group effort should the interest exist. Personally I would love to see an approach that advocates for the leveraging of Canada's own considerable expertise in AI with the need for digital sovereignty, attention to accuracy, lower barriers to access, and the use of open source models to maximize transparency. OTOH, I won't be part of anything that wraps technical issues in culture-war tropes.
(In particular, the open letter's demand for excessive guardrails is at best misguided and at worst laughable. I'm working with local versions of AI models on my own computer as we speak, and it's clear to me that as fast as guardrails can be applied to AI models there are others able to disable them. For a quick example, it's easy to download versions of Chinese models such as DeepSeek which undo all the embedded CCP censorship and give frank answers to embarrassing questions.)
- Evan
On Mon, Feb 2, 2026 at 1:20 PM Marita Moll <mmoll@ca.inter.net> wrote:
Hello advisors. When the invitation to sign an open letter to the Minister of AI (Nov. 2025?), we chose not to sign this letter as some of the wording was considered problematic, specifically it seemed to preclude possible benefits that AI might have to offer. However, the same group has now developed an extensive consultation tool seeking input on the issue of AI implementation to which we could offer our collective wisdom. The e-mail currently being circulated is reproduced below with lots of links to various parts of this project. However, just to simplify, here is a link to the guide and submisison templete which gives you an idea of what this entails.
I think we could offer some useful input. Let me know if you are interested in participating. Deadline for submissions is Mar. 15. As I am probably going to be away from Mar. 3-11, I would need to get this ball rolling very soon.
Maritasee forwarded e-mail below
_______________________________________________ Advisors mailing list -- advisors@tc.ca To unsubscribe send an email to advisors-leave@tc.ca