Artificial Intelligence Everyone now repeats that you should design loops instead of prompting your agents, and at its core that’s true: the loop is the real shift in AI. But these pitches consistently leave out two costs, the tokens that compound with every iteration and the heavier one, sovereignty, because the funnel always ends at a proprietary product to which you hand your keys, your data, and the power to act in your name. This article lays out plainly what a loop actually is, says when it’s a trap rather than a gain, and shows how to rebuild it entirely on your own, on your own infrastructure. Because the only question worth asking isn’t which product saves you the most time, but who owns the loop.
Digital sovereignty There is a humiliation so routine by now that we barely register it: failing, as humans, the very test meant to prove that we are human. We squint at a grid of traffic lights, hesitate over a sliver of crosswalk that spills into the next square, try again, and all the while the bots we […]
Artificial Intelligence On June 16, at its Compile conference, Cursor unveiled Origin, a Git forge designed no longer for humans but for the swarms of agents that push code around the clock. The press saw a GitHub competitor; the real story is the concentration that closes in when a single player brings together the compute, the model, the IDE, the forge, and your working data, a structural risk that would be identical with any other giant. Faced with this single point of control, self-hosting stops being a power user’s affectation and becomes a strategic position, provided you admit one thing: it will no longer be enough to store your repositories, you will have to run your own agent orchestrators or be sovereign but cripplingly slow. The real exit brick, then, is not just Git for the code, it is MCP for the agent, the pairing that lets an open editor restore the same power without handing everything to the dominant owner. My bet for the next three years: not a tipping point but a polarization, in which standardization remains our best way out.
Digital sovereignty There is a sound I have learned to hate. It is not the neighbor’s power drill on a Sunday morning, nor the shrill chime of an incoming email, nor even the synthetic voice of the GPS scolding me when I miss an exit on purpose. No. It is that soft little “bloop,” followed by a […]
Artificial Intelligence On June 12, SpaceX pulled off the largest IPO in history. Four days later it bought Cursor for $60 billion, then unveiled Origin, a Git forge built for agents. The press boiled it down to the purchase of an IDE and missed the point: one man is buying, layer by layer, the right to decide what counts as true, first in code, soon in knowledge itself. The real bottleneck is no longer producing code, it’s arbitrating what deserves to be merged, and for training agents that validation layer is worth more than all of GitHub’s public corpus combined. Yet there’s a crack in the narrative of inevitability: the keystone, Anthropic, is the one piece Musk cannot buy, even as he already supplies part of its oxygen. The real question is no longer to understand his plan, but to decide how much longer we keep renting him the future.
Hardware There is an irony here I find almost delicious, and it is the only delicious thing in this whole affair. The moment my phone becomes unreadable is precisely the one Apple has spent ten years selling me: bright sunlight. Bright sunlight, then. You pull out the device to read a map, frame a photo, glance […]
Digital sovereignty Qobuz has just announced 45.7% growth and a move into the black, a fitting vindication for a French company that bet on uncompromising quality rather than ad-funded free tiers. As a loyal user of its purchase-and-download service alone, I see in it both an aesthetic pleasure and an act of sovereignty. Yet thirteen years after paying for it, the very “Hotel California” that introduced me to the platform is now denied to me for re-download, its rights revoked by the label, its invoice vanished along with it. This mishap raises the real question, that of a “purchase” that is in truth only a revocable access license, and it reminds us that a digital good slips from our grasp the moment an intermediary holds it in our place. The only ownership that counts is the kind you bring home and hold physically in your own hands.
Cryptocurrency A crypto trading bot seduces the developer because it weds technical mastery to the promise of passive income, and that is precisely what should raise a flag. The decorrelation argument has collapsed: the correlation between Bitcoin and equities now peaks during crashes, exactly when you would want it to vanish. The supposed edge of an amateur bot does not exist against algorithmic firms, and paper trading systematically lies upward by ignoring the slippage and taker fees that devour the alpha on every round trip. Even when justified, a crypto sleeve is sized in a risk budget and falls to 1 to 3 percent of capital, a stake utterly out of proportion to the engineering effort a bot demands. That leaves only the honest question: are we after a technical hobby or a wealth-building tool, since automated DCA and rebalancing by API serve real capital where the bot serves only a fantasy.
Artificial Intelligence US export controls were meant to strangle Chinese AI; instead they taught it frugality, and frugality became its pricing weapon. DeepSeek, Qwen, and the swarm of open-weight models now deliver good enough at a fraction of the Western price, which is all it takes to tip the overwhelming majority of use cases. But the market hasn’t tipped where people think: the absolute frontier stays American, and the real moat, distribution, already has the hyperscalers reselling the Chinese commodity on their own compute. For Europe, defaulting to the Hangzhou API means swapping one suzerain for another, when the only real exit, self-hosting open weights, is no free lunch. What remains to be seen is whether the continent will build the conditions, regulatory and industrial, that make this reflex something more than the gesture of an enlightened minority.
Artificial Intelligence On June 8, 2026, Apple unveiled the third generation of its Foundation Models, with one of the most ingenious on-device architectures on the market: AFM 3 Core Advanced stores twenty billion parameters in flash memory and activates only a few of them, routing its experts by prompt rather than by token. Yet the feat is nothing spontaneous, since it is the industrial extension of the 2023 paper “LLM in a flash” that Apple’s marketing would rather leave unmentioned. But behind the on-device brilliance sits a heavier surrender: all five models are co-designed with Google, pre-trained on its TPUs, and the most capable of them runs on NVIDIA GPUs in Google Cloud. The company that had made vertical integration and “designed by Apple” its creed now rents its cutting-edge horsepower from a competitor, precisely where it had promised the most independence. And all the while, the Digital Markets Act keeps Siri AI off Europeans’ iPhones and iPads with no timeline, a fitting illustration of a continent that excels at regulating a match it no longer plays.