Lugh

Silicon computing is starting to run up against hard limits when it comes to energy usage. Bitcoin mining is currently using 2% of the USA's energy. Data Centers are projected to be using a third of Ireland's electricity output by 2026.

However it seems next-generation solutions are on the horizon, and this is one of them. Transitioning computing to energy-efficient new technologies is another front in the war to slow climate change.

When some people see news like this they try and reassure themselves that automation has always created new jobs. You don't see secretarial typists or horse carriage riders anymore, right?

The flaw in this argument is that the AI & robots will be able to do all the new jobs too, but they'll just cost a few pennies where humans were used to getting paid a dollar.

All the people who still think everything is hunky-dory with this and we've nothing to worry about remind me of videos of people on the beaches in 2004 watching the Indian Ocean tsunami coming in, and not realizing until the very last minute how serious things were about to get.

The Chinese automaker BYD reminds me of the famous phrase attributed to the sci-fi writer William Gibson - "The future is already here – it's just not evenly distributed."

Future EV cars will be cheap to own and run. Self-driving tech will lower insurance costs. You can charge them with your home solar setup if you want. They'll last far longer with lower maintenance costs thanks to simple electric engines with few moving parts. As their construction gets more roboticized it will lower their costs further. The batteries that make up a huge chunk of their current costs are falling in price too. CATL, the world’s largest EV battery maker, is set to cut costs in half by mid 2024.

Some people still think gasoline and ICE cars have a long life ahead of them, and don't realize the industries behind both are dead men walking.

Some people are skeptical this technology can ever work, but it appears CASIC's Phase 1 testing in a 2km tunnel has given them the confidence to proceed to Phase 2 testing in a 60km long tunnel.

Chinese railway engineering leads the world so I have a hunch that if any nation can pull this off, then it's China. However, lots of questions remain. A back-of-the-envelope calculation says that to achieve those speeds in the 2km test tunnel deceleration would have been about 3G. That's the same as a rocket at lift-off and not many people's idea of comfort.

AI like Sora won't just affect film & TV industry jobs, but video game production, wider 3D work, and VFX.

I wonder when the paradox is going to sink in with more people. Sam Altman and the like want to tap investors for trillions, yet they are a classic example of a Trojan Horse. Being welcomed with open arms by the very people their actions will destroy. We won't be living in a world with high stock market valuations and investors with trillions to throw around when the creative sector of all nation's economies has become a mass unemployment wasteland.

Renewables are way, way cheaper. Nuclear is finished. I'm sick of hearing its supporters never ending excuses.

They used a powerful machine called a quantum processor to make, for the first time, a brand-new phase of matter called non-Abelian topological order. Previously recognized in theory only, the team demonstrated synthesis and control of exotic particles called non-Abelian anyons, which are neither bosons or fermions, but something in between. Anyons are quasiparticles in a two-dimensional space.

What's exciting is that they can be used to make a quantum computer that is much more stable than current efforts with superconducting qubits.

This is exciting good news for several reasons. Firstly, geothermal energy makes a 100% renewables grid easier, than if that was just solar+wind. As it's effectively on-demand, it irons out the problems with solar+wind's variability.

Secondly, it eases the job transition for fossil fuel workers. Many of the skills and experience in the fossil fuel industry are transferable to solutions like this. Now that the conversation has moved from fossil fuel reduction to setting the timeline for their complete elimination, being able to point to things like this will help win arguments quickly.

It seems like the tech to completely eliminate fossil fuels has arrived, now the remaining issues are political. How can we constantly accelerate and move faster to eliminate coal and oil use more quickly.

Virtual Power Plants are where you use software to integrate a wide variety of power sources to balance large-scale national grids. This means things like home solar, heat pumps, car batteries, and so on.

The US government is working to triple its VPP capacity by 2030. That capacity is equivalent to 80 to 160 fossil-fuel plants that don’t have to be built. VPP's are also a huge boost to the concept of a 100% renewables grid. Objections to that idea usually center around rare use cases when renewables and storage might fail (times when wind and solar have been low for weeks - e.g. extended calm weather in winter). VPP's provide an additional huge source of storage to lessen those concerns.

Computers are starting to use staggering amounts of electricity. There is a trade-off here between the utility of the tasks they perform and the climate damage caused by generating all the electricity they need. Bitcoin mining is thought to be currently using 2% of America's electricity and seems an especially egregious waste of energy.

Radically diminishing computer's electricity requirements as they become more powerful should be seen as an urgent task.

Wow, Feb 15th 2024 will be looked back on as a big day in the history of AI. OpenAi's Sora & this announcement by Google. Both with huge implications.

I've been keeping on top of text-to-video AI as I work with video a lot. Up until now, text-to-video hasn't been able to produce anything impressive, but this is. It's interesting to wonder where this will be in 12 months.