Zuckerberg and Musk Are Fighting About Their Personal Brands, Not AI | Hacker News Reader

Zuckerberg and Musk Are Fighting About Their Personal Brands, Not AI

theatlantic.com

21 pointsmjn10 comments

This article appears to have been removed from the frontpage despite the points to support it.

just making an observation, I had to go back into my browser history to find this when it was recently in the top list and visible.

I think making API calls available for reasons why an algorithm has done this would be helpful in understanding why and how decisions are made.

in relation to this article, I am observing an algorithm making a decision for me that I think I disagree with and I am not delighted about it.

[edit] : to extrapolate on this further, whether either of them is 'right' or 'wrong' I think this is an important conversation to have and design effective means of review now, before we have built complex systems using this foundation and lose understanding of why a machine may make a decision like liquidating all hedges in a hedgefund (read "The Fear Index") or causing a car to turn into oncoming traffic to avoid a deer, being able to have the information readily available for review and improvement is necessary to discuss now... not later.

a horrible analogy I am coming up with now is that the cause of climate change does not matter right now, it is having an immediate impact on people around the globe and we need to take action immediately. (it has impacted my ability to obtain home insurance, I have been denied by every single one except a small one that was recently purchased by geico -- the reason as quoted has been "home is within three miles of the coast line" -- I am on the east coast)

I noticed that as well. I think it may have been manual in this case, going by what I've observed in the past. YC has some idea of what it doesn't want on the front page of its forum, and has paid staff who remove it.

There are algorithms promoting and demoting content too, but afaik they don't have this effect. They either flag posts e.g. if detected as spam (which isn't the case here), or they penalize them, which might push them down to the second or third page, but not remove them from being ranked entirely. A page remaining unflagged but removed from ranking is more unusual, and I think in the past has been manual intervention.

People need adversity to have purpose. I think AI is Musk's way of coping.

One would think he has more than enough adversity on his todo list already.

He has enough adversity managing his risky business ventures. I wouldn't say that his intentions are purely philanthropic, but he has definitely shown that he can grok the fundamentals of a completely different scientific field a lot better than most people can. Zuckerberg was pretty smart himself in his younger days (lol...relatively young), but I'm not sure how much time he spends in the nitty gritty technical stuff v/s big picture business stuff.

In Finance, they call this "Talking up your book." When people make pronouncements, look how it lines up with their economic interests.

I've always just assumed Musk knew something we didn't know. About a stealth project or startup that already developed some form of what could be described as 'general intelligence'.

If not, if he just knows what AI professionals in the field today know, and such stealth project doesn't exist, then I really do think he's worrying about "overpopulation on Mars" as Ng put it. Because there are zero indications in publicly available AI research that we are even close to developing AI that can think for itself, we can't even describe how that would work in theory with any certainty (beyond guessing), let alone implement it.

Weak AI will accelerate inequality in the world. To what degree and how quickly remains to be seen. I think we should be really careful and think of ways to slow down this force. Shame on me for bringing up such matter on a venture capital website...

How so? Musk is someone who's at the forefront of using "AI" for potentially death-bringing self-driving cars. It would be to his disadvantage to call AI dangerous while trying to use it in his cars, just as it would be for Zuckerberg to say that, while using its on his site (which is why he doesn't).

So from that results that Musk isn't in fact doing this for "brand purposes" but because he actually believes AI could be dangerous unless safeguards are put in place, similarly to how someone could say CRISPR is dangerous - unless safeguards are [put in place.

At-least, someone got it

Reply on news.ycombinator.com