MPSimmons

>written in the spirit of sharing with humans.

Not humans who are using AI tools?

I learned this from watching House. I am certain that it's true, but I also kind of wonder how much backpressure there is against this kind of thing because the scanners are also insanely expensive to run per minute.

>I suppose one solution would be to completely vet the training data such that nothing deemed "dangerous" exists in the data, which would be a huge effort.

I can see how this is tempting, but I suspect it would yield a naive model. I think the only way to improve this is to use a model that is legitimately advanced to support the concept of empathy, which may allow it to recognize others as being separate from itself, similar to how toddlers develop this sense (https://blog.lovevery.com/skills-stages/empathy/)

I think it's a side effect of the Transformer architecture. The worldview where all input is equally trusted, and there's no concept of "the other", makes it hard to build effective guardrails where some input is trusted and other input is not trusted.

I didn't say people claimed it was impossible, I claimed that they said SpaceX would never do it. And that's _still_ a pretty common claim.

SpaceX's entire history is full of, "there's no way they'll ever do X", followed by them doing it.

They definitely haven't hit all of their goals, but I don't think anything they want to do is impossible, just really difficult.

It seemed obvious to me that this was bearing stiction and that manually rotating it during the start allowed the fan to spin on its own after that, but I could be wrong and maybe the fan was dead entirely?

That's not necessarily true. Even spaceships in LEO will perform temperature-driven rolls so as to distribute heat and radiation. I have to assume that long-term ships like interplanetary transport will do the same.

Professor Hubert Farnsworth: Dear Lord! That's over 150 atmospheres of pressure!

Fry: How many atmospheres can the ship withstand?

Professor Hubert Farnsworth: Well, it's a space ship, so I'd say anywhere between zero and one.

I've considered whether our current transformer-based AI could be conscious, as I understand it, which I deem to include some degree of self awareness combined with some degree of external awareness. I can see how theoretically something could be self aware without any external awareness, but I grasp at straws when I try to envision what that experience could be like.

In either event, I think transformer-based AI can only be conscious during the act of inference. If that's the case, then the experience of consciousness that the AI is subjected to must be the content of the tokens in the context window and the activated weights. Maybe that's reason enough to be polite to our agents?

I knew it was going to be Patrick Boyle before I even clicked.

Oh, I'm interested - do you have any docs with human responses to that?

I worked at SpaceX at the time, and I cannot speak for the company, but I can tell you that approximately nobody inside SpaceX took the idea of a sniper seriously. There was a lot of internet talk about it, and it was one of hundreds of avenues that were explored, and ruled out basically as soon as it was explored.

The very interesting part of the liquid oxygen failure (and this was published in the investigative findings) was that the liquid oxygen that became trapped in the fibers was actually cooled and compressed into solid oxygen - you can read some details here: https://www.americaspace.com/2017/01/02/spacex-closes-amos-6...

It was very likely the largest explosion in Florida spaceflight history. Considerably larger than when SpaceX blew up AMOS-6 in 2016, and that required a full rebuild of the pad infrastructure over 18 months.

Intentionally malevolent is kind of their thing in this administration

Consider for a moment the data requirements for the telemetry system that records those engine runs.

"Since the beginning of time, man has yearned to destroy the sun."

What is the advantage of the screw press? From an outsider's perspective, it _sounds_ slower?

I was fortunate enough to be chosen to talk to astronauts on the ISS from the ground and ask them a question, and as far as I can tell from searching this awesome resource, I think my question was unique.

I asked Ann McClain and David Saint-Jacques what experiment or module they would add to the ISS if they could pick anything. They both agreed that a rotating wheel that simulated gravity would be helpful for experiments and quality of life.

I think they _could_ but I doubt our current activation functions are sufficiently nuanced to allow consciousness that we would recognize.

It reportedly has a $2B ARR, and a 5x multiplier doesn't seem insane to me, but who knows, honestly

So, John Apple?

THEORETICALLY the government has _some_ restrictions against monitoring US citizens (https://www.brennancenter.org/our-work/research-reports/soci...), and it sounds like none of those are deemed to apply if you're using a VPN, per the original article.

Yes, Elon's trajectory has been absolutely unbelievable since 2019 or so. Talk about a guy who cratered his whole reputation. It's a shame.

Difference in cost per launch, cost per mass, or payload capacity?

> it’s not quite correct to suggest there’s a vibrant external market for their product

There is a very large demand for launch services. SpaceX balances launching customers and launching Starlink. It's not like they give every launch slot to customers and then launches Starlink whenever there's an opening they couldn't fill.

SPF or DKIM maybe?

I don't disagree with the sentiment, but I think the signals that we use to determine whether we're doing the right things are different with the new AI enhanced toolsets.

Refactoring decent sized components are an order of magnitude easier than it was, but the more important signal is still, why are you refactoring? What changed in your world or your world-view that caused this?

Good things still take time, and you can't slop-AI code your way to a great system. You still need domain expertise (as the EXCELLENT short story from the other day explained, Warranty Void if Regenerated (https://nearzero.software/p/warranty-void-if-regenerated) ). The decrease in friction does definitely allow for more slop, but it also allows for more excellence. It just doesn't guarantee excellence.

> give them parachutes at least

the first time

Honestly, they're probably subsidizing Elon already via Tesla, but the super disturbing part here is what the author nails when he says the tail is wagging the dog. Indices should reflect market investment, they shouldn't drive it like this.