madrox

5 ms·

31.
by madrox 1mo ago
Ever heard of an MBA?
32.
by madrox 1mo ago
Not sure why you got downvoted. You're mostly right.
33.
by madrox 1mo ago
Happy to hear someone else say this. Similar story. I could never live in SF again but roaming around it was nostalgic in a way going back now just isn't
34.
by madrox 1mo ago
In addition to what another reply has said, I'll just say any company that does not make me feel like they'll ban me if I hold their model wrong. The bar is low.
35.
by madrox 2mo ago
Anthropic lost all good will with me. Everything from their policies to their rhetoric has an air of "we don't trust you." The way they treated users who wanted to use them for OpenClaw didn't sit well with me, and then
36.
by madrox 2mo ago
I applaud GitHub. However, I think no matter how valiant they are they will not climb out from under this. The scale problem will keep getting worse, and it's getting worse in a way I don't think is translating to more money for t
37.
by madrox 2mo ago
I think "the pseudocode is persisted alongside the generated code" just reinvented jira/linear tickets and PR descriptions. We have ways of using git and tracing the code write to the thought process behind it.
38.
by madrox 2mo ago
I'm not normally cynical to such things, but I have a hard time taking this pause justification at face value. It has too many convenient side effects, and chief among them is cost savings. There's a new wave of warnings that the
39.
by madrox 2mo ago
I think the difference between your opinion and others is that for many being a dad is the intention and not an expectation. They want the experience of being an active and present parent. If it were purely about obligation, in this day and
40.
Engineers will do anything to avoid learning from history(horn.gg)
178 points by madrox 2mo ago|117 comments
41.
by madrox 2mo ago
I think it funny how much average engineers are beginning to discover the challenges of engineering leadership and program management. This has always been the bottleneck. It's why managers and PMs want to be in standup. It's why
42.
by madrox 2mo ago
A warrant is not required for an AMBER alert, which amounts to poor man's mass surveillance of license plates. We're basically fine with an incredibly intrusive push notification to get people to do the work without hand-wringing
43.
by madrox 2mo ago
As trying things go, it doesn’t seem that crazy of an idea. I imagine air traffic control would even make for a pretty fun video game.
44.
by madrox 2mo ago
I don’t find that funny at all. AI didn’t invent the problem of balancing profit motives with ethical guard rails. I wouldn’t be surprised if many AI ethics responsibilities move to legal in the future.
45.
by madrox 2mo ago
I am not in the AI ethics field, but I've watched multiple tech-adjacent disciplines that say "that's what engineering and product are for" get subsumed by engineering and product, who don't really care for a group
46.
by madrox 2mo ago
This reminds me of when systems administration rebranded to devops
47.
by madrox 2mo ago
I think it interesting that Bakalar was at Meta for six years before joining OpenAI. A lot changed across the industry in those six years, and in the meantime I wouldn't be surprised if all Bakalar learned was how to play Meta politics
48.
by madrox 2mo ago
Maybe if I put it that way there'd be less controversy in the replies
49.
by madrox 2mo ago
I find that hard to believe. Between AI doomerism and the drama at Google years ago between some researchers and Jeff Dean alone you can get at least 500 pages.
50.
by madrox 2mo ago
I don't really have any particular disdain for AI ethics. I think the field is changing from one of high theory to high application. People good at one are not usually good at the other, and says nothing about OpenAI's view of AI
51.
by madrox 2mo ago
I don't see either of those as reasons not to train on it. Training isn't zero sum. You don't need to take something out to train ethics, and whether you explicitly train it or not, models will take ethical stances. Your seco
52.
by madrox 2mo ago
I agree, and I don't think many first wave AI ethicists would be on board with that, which is why the field is in for a rude awakening for the next few years if not already.
53.
by madrox 2mo ago
> What does AI ethics even mean? This, I think, is the question at the core of of the field right now. Five years ago it was highly hypothetical. Today, not so much. I'll be curious to see what happens.
54.
by madrox 2mo ago
I agree, which is a different job from what I expect most AI ethicists imagined it would be.
55.
by madrox 2mo ago
I was specifically thinking of Candide. A bunch of coddled people thinking we're in the best timeline suddenly subjected to harsh realities.
56.
by madrox 2mo ago
Because leaving that to emergent behavior is silly. Why is the idea of training models on ethical reasoning so strange?
57.
by madrox 2mo ago
The ethicists who see their job as publishing papers and acting as philosophy professors, sure, but someone has to own the training and eval suite for how AI thinks about trolley problems.
58.
by madrox 2mo ago
"Prove" is a shorthand because of course it's a stochastic process. My point is that I would expect ethics to be part of the training and eval criteria.
59.
by madrox 2mo ago
Microsoft Tay has entered the chat. I'm kind of surprised by the cynicism. There are real problems in AI ethics that contribute to model training. Someone has to own those problems. How that team is incentivized is outside the scope of
60.
by madrox 2mo ago
In five years, I'd love to read a book about the history of AI ethics. I suspect it will read like Voltaire. My sense is that it is radically shifting from a fluffy marketing arm to a department expected to contribute meaningfully to d

More ›