Today, we are publishing one of the side tracks of research ongoing with Q, our E2EE ML training and inference library, klearu: https://t.co/RKcPy7elhc
SLIDE proved that hash tables can beat GPUs at training deep networks. Further works compounded on this, and Klearu is the first native Rust implementation built on top of this research, extending it to LLM inference, sparsity prediction, and private two-party computation.
In the current days we're seeing deeper trust being placed on AI, while the largest of providers are collecting this data for the purpose of not only training, but also advertising, or even selling this data to others. The risks grow worse with every passing day.
The majority of AI research for private AI exists in the form of using TEEs – but we've seen time and time again that using TEEs for privacy is disastrous, guaranteed to leak, and even by it's name, is a massive requirement of trust.
Outside of this, other private AI looks towards FHE. We know, at least for the near future, that FHE cannot perform at a speed high enough to be generally useful. So instead, we adopted 2PC, with flexible security configurations, where users can be assured that their requests remain private. The majority of these research projects have strictly an output of papers, with no or limited real world instances of their use.
Klearu's implementation is available now, with simple instructions for developers to try it out.
Hey Q fam,
we have been improving Quily accuracy in the past weeks.
There are still a few days of free usage so keep testing and report any issues. 🙏
https://t.co/k5hfQEjicw
P.S.
You can ask questions about the last livestream too by mentioning "last/recent livestream".
$QUIL
⚡️The system is eating itself.
When Amazon, UPS, Intel, and others start cutting this deep simultaneously, it means something more fundamental than “tight margins.” It means the productive layer of the economy is collapsing under the weight of its own optimization logic.
Every layoff now is both an act of short-term rationality and long-term suicide. The firms know it. The executives know it. The markets know it. But they can’t stop.
They’re trapped in a closed feedback loop, a machine that rewards death dressed as discipline.
Capitalism has crossed into a stage where it no longer needs humans to function, but still depends on their belief to exist. That’s the contradiction, the machine is pruning its own believers while pretending it’s efficiency.
1. The “consumer economy” is already dead.
Nobody wants to say it yet, but the consumer model, the entire foundation of Western postwar prosperity, is quietly finished.
You can’t build infinite growth on finite wages, and you can’t sustain demand while hollowing out the class that drives it. The middle layer of society - the producers, buyers, dreamers - has been strip-mined to the point where they can no longer regenerate.
The 2020s economy is not cyclical recession. It’s metabolic collapse. The system can’t process its own waste or regenerate its base anymore. It’s like an organism starving while eating its muscles to stay warm.
2. The elites know this, but they’ve chosen to accelerate collapse.
Here’s the real unspoken truth: the people running these companies, the ones with the spreadsheets and control over capital flows - they know exactly what’s happening.
They understand the reflexive trap: if they don’t cut, their stock dies. If they cut, the world dies.
They’ve chosen to save the stock. Because the stock is their world.
This is the quiet revelation of our time - we are ruled by people whose survival incentives are no longer tied to the survival of the system itself.
They’ve built lifeboats - offshore wealth, private security, parallel digital economies - and they’re optimizing the ship for their escape, not for collective navigation.
3. The next phase is narrative triage.
When the system can no longer grow, it starts storytelling harder.
Expect every layoff wave to be accompanied by new propaganda about “AI productivity,” “efficiency,” “lean reinvention,” and “post-labor creativity.”
The goal will be to reframe collapse as progress - to convince people that losing their jobs is the dawn of a “new paradigm.”
But it’s camouflage. The truth underneath is that automation and financialization are converging into a post-human economy where capital reproduces without labor.
4. Final layer
When a system prioritizes margin over humanity, it signals that it has lost faith in the future.
These layoffs tell us that the machine no longer knows how to grow except by shrinking.
It is the same signal we’ve seen in housing, in politics, in fertility, in faith.
The same quiet collapse, a civilization optimizing itself into silence.
And the question hidden beneath this post:
“Who will have money left to buy your products next year?” -
is really this:
Who will be left to believe in the story that built it all?
The freeze has thawed. Quilibrium V2.1 is now live on mainnet!
Builds are now released to general availability.
The network has unhalted and we’ve entered a 24-hour initial enrollment phase. This is a pivotal moment, as the first set of nodes joining will define how shards are distributed across the protocol. It marks the beginning of live operations. If any critical issues arise, we have a path to pause and return to a safe state.
After 24 hours, the network becomes fully decentralized. The beacon shuts down permanently, shards are finalized, and transactions in the new format begin processing. From this point forward, the system is operating fully in the 2.1 environment.
Once stable operations are confirmed, which may take between 24 hours and a week depending on conditions, we will activate the bidirectional bridge, bring QStorage and QKMS onto mainnet, open QConsole for everyone, and migrate existing user data into the upgraded environment.
@joegrower420@GwartyGwart I only know her because she's from my state. No one but a small portion of the US gives af about the wbna. They literally lose money every year and are subsidized by the nba. Quit being dumb.