Reminds me of this post[0] from a few weeks ago:

>A couple years back, I got a job offer from an investment bank to help them win zero sum games against people who didn't necessarily deserve to lose. I had tried very hard to get that offer

https://www.hgreer.com/PlayingInTheCreek/

>Many competitive people want to win, but they want to win the real game, not a rigged version.

Apply the phrase to the staff member he lost to, and the situation makes sense. The staff member wants to win the real game (of remaining a high-salary Facebook employee), and will throw an otherwise inconsequential game of Catan to maintain that position's security.

I have some incomplete thoughts that the rise in LLMs is in part driven by society's willingness to accept half-accuracies in a post-truth world.

If the societies of 2005 had the technologies of 2025, I expect OpenAI/Anthropic etc would have a much more challenging time convincing people that "convincingly incorrect" systems should send Nvidia to a $1tn+ valuation.

This sycophant-as-a-service feature is already close to the way the major LLMs currently work. Discuss any moderately controversial topic with them, and they'll lean into your opinion within a couple of comments.

There are closed-loop datacenters, but the hyperscalers now use evaporation chimneys. Google used ~21bn litres of water in 2023-24.

Doesn't that fall back down again as fresh rainwater?

Sure. While lowering the recharge rate on the aquifer, the river flow, etc. Instead of being re-used downstream, it's re-used downwind. Snow melt original destined for the over-allocated Colorado might instead end up in the Mississippi.

All that irrigation water pumped out of the ground in the southwest falls back as rain too. While the land subsides, plants can no longer reach water, and towns have to build deeper wells.

This is a Google Workspace site thrown up 11hrs ago, and doesn't appear to be linked to from any official source.

I don't think it's credible that CVE as an organisation would produce this website and not link to it from their official site or social media accounts.

It's got all the hallmarks of a piece written by AI. Lots of purple prose, adjectives where they add no information, bullet lists, etc. All this sits alongside banal content like "[Wiz] Use colors and designs to signal reliability in a high-stakes industry."

It may have been lightly edited & enriched by a human, but most of this article was written by an AI.

In my experience, what's being made up is an incorrect name for an API that already exists elsewhere. They're especially bad at recommending deprecated methods on APIs.

>The idea is to think outside the box and create entirely new markets instead of just new products in existing ones

It's interesting that SV outwardly says it "wants to create entirely new markets instead of products in existing ones", meanwhile the actual experienced outcome for users is the same experience across multiple markets.

SV is somehow failing on both of its metrics here. It's creating entirely homogeneous products across all existing markets.

By "create new markets" they've always meant "Become useless middlemen by displacing the existing bridge between makers and consumers"

Usually their new bridge is modestly more convenient in some way, but opens the door to the worst kind of enshittification.

>There are plenty of us that think that dramatically expanding the power of copyright would be a huge mistake that would primarily benefit larger companies and do little to protect or fund small artists.

The status quo also primarily benefits larger companies, and does little (exactly nothing, if we're being earnest) to protect or fund small artists.

It's reasonable to hold both opinions that: 1) artists aren't being compensated, even though their work is being used by these tools, and 2) massive expansion of copyright isn't the appropriate response to 1).