Yes, it's marketing straight out of the Big Tobacco playbook. Convince everyone your product is dangerous, get government to ban it, and then... uh... pivot into adjacent market segments?

> If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers.

... Isn't that basically what Anthropic asked for, literally a week ago?

https://www.anthropic.com/institute/recursive-self-improveme...

> We believe it would be good for the world to have the option to slow or temporarily pause frontier AI development to enable societal structures and alignment research to keep up with the advance of the technology.

I don't understand why headlight elevation isn't standardized.

It was. We had sealed beam headlights for a while till we didn’t. There were common rules for aiming and it worked. The lights weren’t all that bright and the styling was not stellar, however.

I remember having to take my car in to adjust the aiming of the headlights after it didn't pass inspection. So we used to take things like this seriously. I just had my car inspected last month, I don't even remember them hitting the horn. I'm guessing they pretty much just shove a sensor up the exhaust pipe and call it a day while accepting your payment.

No shoving sensors required, the data is all in the ECU accessible over OBDII interface. The car knows if it’s compliant in real time using the sensors it already has.

Wouldn't the VW dieselgate show that we can't trust the manufactures? A third party sensor would be more trustworthy to me.

In states in the USA which perform emissions testing, many of them did not mandate it for diesel cars. For example, I owned a VW Jetta TDI and in New York (which has yearly emissions testing where an OBDII computer is mandated to be connected to gasoline powered cars in order to pass the yearly emissions inspection) and I was exempt from the emissions testing entirely.

A 3rd party sensor would be incredibly expensive for inspection stations to purchase as it would need to meter the air and fuel which enter the engine (assuming we aren't going to trust the car's computer which already knows these figures) as well as to measure the emissions out of the tail pipe. This is economically unrealistic to implement without a dramatic price increase in the cost of regular emissions testing.

Trusting the computer is the economical and realistically widely implementable solution. But yes, it has it's blind spots.

That sky crane system still blows my mind. Straight out of science fiction.

Assuming https://en.wikipedia.org/wiki/Sky_crane_(landing_system) can I ask which part of the system did you find most impressive? https://www.youtube.com/watch?v=p1WX0CATyn8

Mainly because it's a complex multi-step plan with so many potential failure points, which would already be impressive if executed on Earth but even moreso being executed on a distant planet where the conditions are different enough that it can't be fully tested in advance, and yet despite that the whole scheme worked flawlessly.

But if I had to name a specific part, I'd pick the control system. The skycrane is dangling a heavy rover from a pendulum controlled by rockets. It's unstable in every axis and has tight performance requirements to let the rover down softly and not kick up dust. Just very impressive.

My understanding was the 20m tether length was designed to avoid dust. More interesting to me were the choice of nylon vs. other polymers, the equal-release multi-line spool design and the choice of cutting the lines on the rover (permanent mission-long mass penalty) vs. skycrane (which is discarded).

It worked out. Philae lander with harpoons was comparably complex but mostly failed unfortunately.

The audacity.

I agree that AI poses a threat to society. I act on this by not developing world-leading AI models and offering them to anyone willing to pay top dollar, while funneling that money back into accelerating AI capabilities development. Maybe Dario would consider taking a similar ethical position? Maybe he would support restrictions and taxes on data center construction, in order to slow down the pace?

If Anthropic were not developing these models, one of many other companies would be. I think it's good that the CEO of the current world-leader is at least considering these discussions and platforming possible solutions.

The fact that he doesn't support more restrictive approaches that don't align with his incentives doesn't invalidate the points he is making.

> If Anthropic were not developing these models, one of many other companies would be.

"Everyone else is doing it" but even worse as nobody else is actually even doing it and you just made up that part.

Just commenting on this horrible justification trick in general.

And hopefully the other candidates applying for the position have similar inflation expectations to your own!

Surely the variable resistor would only have been on the field winding. It wouldn't waste that much energy.

Figuring out where to place it, or how to do this took time. I'm sure some early attempts were really bad.

Synthetic inertia has come a long way. An inverter knows just as well as a rotating generator what frequency and phase it should be generating, because it can mimic exactly the ideal equations of motion of a rotating generator with programmable inertia.

The government may have allowed it with that intention, but the corporate leaders followed through mainly with the intention of short-term share price increases. I don't see how the same incentive isn't in place today with respect to data. Perhaps only the perception of China's ability to outcompete its American customers has changed.

I think Tokugawa-era samurai fits the description better. The Meiji era saw the samurai stripped of most of their stipends and privileges, and with little left but their pride they had to go find respectable jobs.