Profile avatar
bretthmeyer.bsky.social
We live because everything else does. —Richard Wagamese He/him, Professor of ECE, researching hardware-software co-design of machine learning systems; 🇺🇸 in 🇨🇦! Views expressed are my own. https://rssl.ece.mcgill.ca/
228 posts 308 followers 443 following
Regular Contributor
Active Commenter
Posts 24 Comments 25

AI is a much deeper issue than just the classic "more technology is more good" that the tech industry has promoted for the last few decades. It's an express goal of AI companies to destroy the livelihoods of dozens of careers (e.g. artists & musicians). Why wouldn't people be angry? Thread ⬇️

submitted 10 days ago • 3 comments

Good morning, beloved. Perfection is not a prerequisite for good. Keep trying. Today is a great day to overcome evil with good.

submitted 12 days ago • 1 comment

I’ll have my work cut out for me as I try to convince my computer engineering students to do the work to gain expertise, and not simply delegate the hard work of thinking to cognitive automation tools.

submitted 16 days ago • 1 comment

PhD Timeline xkcd.com/3081

submitted 18 days ago • 614 comments

“Nearing?” GTFO.

submitted 25 days ago • 0 comments

DOGE is operationalizing one of the harms of the term "AI". How you describe your science matters. buttondown.com/maiht3k/arch...

submitted 35 days ago • 3 comments

The erosion of trust in science & scientists — sometimes due to complex social/historical factors but often deliberately perpetrated by charlatans — has tragic, deadly consequences, and, if unchecked, is an existential threat to society.

submitted 63 days ago • 43 comments

From a poetry collection by Mary Oliver, where after a hundred poems showcasing gentle observations on nature and animals, she hits you with this

submitted 64 days ago • 191 comments

It’s astonishing to me just how reasonable they all thought the Trump administration would be. apple.news/A1_gj535SS_G...

submitted 75 days ago • 0 comments

to budding scholars out there: your scientific paper is as good as your ability to present a clear and coherent storyline. without your compelling narrative, i have no reason to care about your data or findings

submitted 77 days ago • 2 comments

Stand by this: www.politico.com/newsletters/...

submitted 83 days ago • 87 comments

As I prep to teach Computer Architecture again next year, it’s clear that I need to prep students to question the veracity of LLMs so they’ll put in the work develop their own expertise. “Let’s ask ChatGPT” will be returning!

submitted 84 days ago • 0 comments

this might be the case for *some* aspects of *some* jobs alas, if you don't have solid domain expertise knowledge to verify llm output, you're done

submitted 84 days ago • 8 comments

Your imposter syndrome should have died laughing at this point. You're doing great. You didn't author a bill to rename Greenland "Red, White, and Blueland." You can do anything.

submitted 90 days ago • 19 comments

One of the things I love about my office is the student-society-operated ice cream shop on the ground floor of my building. Ice cream is: cheaper on “Toonie Tuesday;” free if you’ve failed an exam; and, discounted in general if you bring your own cup! 😍 Spontaneous ice cream breaks FTW!

submitted 96 days ago • 0 comments

Ugh #GT7, putting dirty drivers (DR B, SR C) in the same lobby as clean drivers (DR C, SR S) is soooo frustrating.

submitted 100 days ago • 0 comments

No, I do not think we should use gAI to build literature reviews, and we definitely shouldn't be teaching faculty how to do this, this is not just some fancy new search engine, what on earth is wrong with you and why are you doing this ffs

submitted 102 days ago • 32 comments

Yes, yes, yes, and yes.

submitted 102 days ago • 0 comments

Me to my achy legs, as I ramp up into another marathon training cycle: Foam rolling will continue until morale improves.

submitted 103 days ago • 0 comments

👉 DeepSeek may well have broken OpenAI’s Terms of Service and distilled their intellectually property without permission 👉 OpenAI may well have done analogous things to YouTube, New York Times, and countless artists and writers. 👉 Karma is a bitch.

submitted 104 days ago • 53 comments

Hell hath no fury like an uncaffeinated researcher.

submitted 105 days ago • 0 comments

Why do people who seem otherwise intelligent and competent keep suggesting GenerativeAI as a solution to problems that require accurate information when the one thing it can’t deliver is…accurate information?

submitted 105 days ago • 143 comments

As someone who has reported on AI for 7 years and covered China tech as well, I think the biggest lesson to be drawn from DeepSeek is the huge cracks it illustrates with the current dominant paradigm of AI development. A long thread. 1/

submitted 106 days ago • 227 comments

Rude. What did Pluto ever do to you, Hank Green?

submitted 109 days ago • 0 comments