AI Developed Rocket in 2 Weeks

[JoeSchmuckatelli]

 

"rather than the traditional process of countless engineers spending months, even years manually engineering a model in programs like CAD, this one was designed in two weeks—thanks largely to AI." 

Copper seems initially like an odd choice for a rocket engine given its low melting point, but apparently it enables "compact high-performance engines" when actively cooled. 

The engine is designed to produce 5 kN of thrust (equivalent to 500 kg/1120 lbs of lift mass or 20,000 horsepower), and Leap 71 says it would be suitable for the final "kick stage" of an orbital rocket

https://www.pcgamer.com/hardware/this-20000hp-ai-generated-rocket-engine-took-just-two-weeks-to-design-and-looks-like-hr-gigers-first-attempt-at-designing-a-trumpet/

... All PC Gamer rocket news is good PC Gamer rocket news 

 

Edited June 21 by JoeSchmuckatelli

[K^2]

Copper is a good target material for cheap sintering, so I get why it's desirable, but I'm impressed that it works. Even with cooling, I'd expect engine-rich exhaust from this thing. But hey, if this is good enough for one launch of an expendable craft and is cheap to print, I don't think anyone would complain.

[DDE]

13 hours ago, JoeSchmuckatelli said:

Copper seems initially like an odd choice for a rocket engine

* laughs in bronze *

13 hours ago, JoeSchmuckatelli said:

AI

Designed autonomously "without human intervention", huh? Sorry, even before seeing the mention of UAE this stinks "investor bait" almost on the level of Theranos, minus the immorality of medical fraud.

Edit: after having slept on it, let me refine my point. It's possible, even likely, that the statement is technically correct and that, after the humans meticulously defined all parameters, an iterative design algorithm took two weeks on a supercomputer to optimize the engine. I'm doing the wrong thing of considering only the context and the extraneous circumstances.

UAE is swimming in money - not just oil, "Dubaysk" is a major destination for Russian capital, UAE isn't too stringent on sanctions enforcement and AFAIK almost ended up on FATF's grey list of jurisdictions with lacking anti-money laundering controls - and has a political mandate to diversify its economy. This means it can and will throw money at anything with "space", "AI" or "nano" like it's 2018. Saudi Arabia spends money on architectural abominations like The Wall, UAE spends it on space. Good for them, but we must acknowledge the Middle East is a major magnet for very shifty and outright fraudulent start-ups, and that the one under discussion has red flags all over.

Edited June 22 by DDE