Data center emissions likely 662% higher than big tech claims. Can it keep up the ruse? Emissions from in-house data centers of Google, Microsoft, Meta and Apple may be 7.62 times higher than official

Shocker, tax avoiding megacorps use the same accounting principles for carbon avoidance. Presumably with te aid of the same in-pocket politicians.

At this point, my only hope is that the failed business model will kill most of these before they get even worse, regulators don't seem to have any appetite to monitor or restrict/tax the runaway energy usage.

You don't say?

are likely about 662% – or 7.62 times – higher than officially reported.

This annoys me. If the value is 662% higher than reported, then it is also:

• 6.62 times higher than reported
• 762% as high as reported
• 7.62 times as high as reported

Whether you use percentage or a factor doesn't change the principle. "as high" signals a new value that is a straight multiple, "higher than" signals a change that is a multiple, so always assume a base of 1 or 100% on which the change is added.

By the way 1 and 100% is the exact same thing, and so is 662% and 6.62. Percent aren't magical they are just a short form for saying hundredths of 1.

In everyday speech they also signal that you're talking about a fractional value in relation to a different one that is hopefully clear from context. But it isn't always and people abuse that too.

It seems to boil down to the question of how much you believe in the Recs market working.

The way I understand those Recs is, they pay extra to have green energy produced, but it's not in the location where they are actually using energy. So their actual energy use is much more grey than what they pay for, but they essentially say as long as the correct pricing signal reaches the location independant energy market this is good for green energy buildup.

Now the idea behind that must be that it's more sensible to support green energy production in places where projects are being built, where they perhaps can be most efficiently built, rather than trying to bribe the local suppliers into building green energy locally, where perhaps the location isn't even well suited. It's kind of understandable from that perspective.

But I can see how there would be lots of corrupting influences here. The companies local to the green electricity producers can factually claim that the local power is greenly produced without paying for the Recs, and the remote companies can pay the Recs and claim they are supporting green energy. In the end you have a sort of double counting of the green energy, swallowing up some of the market demand for an actual buildup of more green energy capacity. Or even more nefarious, maybe a company could start issuing more Recs than green energy production actually justifies.

The system also masks out the issue of actually transmitting and storing the green energy from where and when it's produced to where and when it's needed. Even if you have perfect Recs accounting in the system, nothing guarantees that the pricing signals align time wise. Green energy prices should be higher during the night or winter, but that's not reflected in time independant Recs.

And finally the existance of the Recs system absolves the data center builder from choosing a location where green energy can be produced or transmitted to. If this ends up placing a datacenter close to a coal power station, and then they buy Recs from a solar park in a desart far away. In the end this makes it harder for the local community to overcome the need for the coal power station. It might have been possible to build enough local green energy and storage without the datacenter next door, but once it's there soaking up lots of local capacity it's going to be much harder to replace the local grey producer.