Building shared machine intelligence at https://xitter.cf/t.co/ez0CmsBUcm Optimizing human health at https://xitter.cf/t.co/cWEyn6QRHP orchestrating life on Mars at https://xitter.cf/t.co/YXbeOYdXSq
Kardashev Shed
Joined January 2013
- Tweets15.4K
- Following2.9K
- Followers1K
- Likes141K
Pinned Tweet
There's currently no way to make asphalt on Mars, or the moon, or asteroids. A reasonable solution to this is to melt pools of the regolith and blade the liquid (primarily basalt) together with the abundant aggregate. This has been explored by the ESA in 2022 and It's time to try again.
populationmars.com/construct…
Previously I built a simulator for laser ablation of orbital debris fields, a plan to avoid Kessler syndrome. Here's how these lasers could be used to modulate the flight paths of something like the Russian COSMOS 2612 after it began maneuvering dangerously close.
This isn't 'space war' per se. There would be some damage, but it's possible to move an improperly pathed object without completely destroying it. This may prove necessary for valuable but dysfunctional satellites that are dangerous but recoverable. Here's the simulator: orbitfleet.populationmars.co…
The Russian satellite Cosmos 2612 is flying insanely close to a US licensed defense security ICEYE satellite. While it's likely no real damage occurs, this cat and mouse game has big implications for dual use space suppliers and the space economy as a whole. populationmars.com/blog/the-…
Plenty close enough for jamming Xband or spoofing GNSS, unlikely though. There’s lots of redundancies in an active combat SAR system. Spying to figure out exactly what to jam on the ground is more plausible.
Benjamin Cox retweeted
The ideal outcome for science isn't a couple of labs using proprietary models and millions of dollars to crack trophy problems, it's to have millions of AI-aided human scientists who follow their curiosity and become way more ambitious
I did a swarm alignment project. Basically it works by using a type one classifier (jev) to spot misaligned agents (and their artifacts) in a swarm and culls them before they can lobby other agents. Then another agent is respawned.
This was a weekend project and there's a lot of threads to pull on.
- Training a custom type one classifier instead of jev
- Inducing the misalignment rather than seeding it
- Attempting classification on a 'no coordinator' swarm
- Training for steganographic collusion.
If anyone wants to contribute compute for the poor, or just hack away on the concept with me dm's are open.
Benjamin Cox retweeted
There's currently no way to make asphalt on Mars, or the moon, or asteroids. A reasonable solution to this is to melt pools of the regolith and blade the liquid (primarily basalt) together with the abundant aggregate. This has been explored by the ESA in 2022 and It's time to try again.
populationmars.com/construct…
Benjamin Cox retweeted
Technical moats are mostly over at this point. If you’re building something that is meant to be durable, the only moats now are:
• Proprietary data
• Network effects
• A physical component
