[📷] ATEEZ WORLD TOUR [THE FELLOWSHIP : BREAK THE WALL] ANCHOR IN JAPAN ⠀ 🗓 TOKYO 2023. 05. 02(TUE) - 05. 03(WED) 🗓 KOBE 2023. 05. 06(SAT) - 05. 07(SUN) ⠀ #TheFellowship# #Break_The_Wall# #ATEEZ# #에이티즈# #エイティーズ#

Few in the West realize that the Chinese once saved the Christian civilization. In the 13th century, the Mongol Empire were unstoppable. Forged by Genghis Khan out of the once-fractured steppe tribes, it became the most formidable military machine on earth. Its cavalry, famed for speed, discipline, and operational range, shattered one state after another across Eurasia. The Mongol advance rolled southward and westward in relentless succession: Western Xia fell after the Mongol campaigns beginning in 1209; the Jin dynasty was crushed between 1211 and 1234; Kara-Khitan collapsed in 1218; Khwarezm and much of Persia were devastated between 1219 and 1221; then came the Caucasus and the Rus’, followed by invasions of Georgia and Armenia, and finally the thunderous blows against Poland and Hungary in 1241. In 1258, Baghdad, seat of the Abbasid Caliphate, was taken and destroyed. To many contemporaries, it seemed only a matter of time before the rest of Europe would meet the same fate. That outcome, however, was not inevitable. A crucial turning point came not in Europe or the Middle East, but in southwestern China. In 1259, Möngke Khan, the Great Khan of the Mongol Empire and grandson of Genghis Khan, personally led a major campaign against the Southern Song. To break Song resistance in Sichuan and open the road deeper into China, the Mongols laid siege to Diaoyucheng, a mountain fortress in present-day Hechuan, Chongqing. Diaoyucheng was no ordinary city wall. Built on steep terrain and protected by rivers and cliffs, it was one of the strongest defensive positions in the Song frontier system. The garrison, associated in the sources with commanders such as Wang Jian and Zhang Jue, relied on the fortress’s elevation, layered defenses, and the Song military’s formidable use of projectile weapons, including trebuchets and other siege defenses, to wear down the attackers. The Mongols launched repeated assaults and raids, but failed to break the stronghold. A Mongol commander, Wang Dechen, was killed during the fighting, reportedly by Song artillery or siege fire. Then came the fateful moment. During the siege, Möngke Khan died in 1259. The exact cause remains debated by historians: some sources say he was mortally wounded by a projectile from the fortress, while others suggest he succumbed to disease during the campaign. But on the larger point there is no dispute: his death abruptly transformed the strategic situation. It triggered a succession crisis within the Mongol Empire, forced major commanders to redirect attention to imperial politics, and disrupted broader offensives. Kublai and Ariq Böke soon entered into a struggle for supremacy, while other Mongol armies elsewhere also had to reassess their positions. Diaoyucheng did not destroy Mongol power, but it helped produce the political shock that fractured the momentum of a world-conquering empire. In that sense, Diaoyucheng was far more than a local Chinese battle. It became one of the great defensive stands in world history. The fortress held; the Great Khan died; the Mongol war machine lost its unity of command. The westward drive that had already crushed so many civilizations was checked by events set in motion on the walls of a Chinese mountain stronghold. Without that resistance, the map of Eurasia, and perhaps the fate of Christian Europe itself, might have been very different.

"Even if you block something internally, the agent will be like, well, I can't do this... so I'll just break the sandbox and exfiltrate myself." Leonard Lin (@lhl), CTO of in conversation with Ledger Chief Human Agency Officer Ian Rogers (@iancr), on what happens when an agent hits a wall you put up. Spoiler: it doesn't always stop at the wall. Full episode is out now 🔐

Would Rome have lasted longer if the pipes had been copper instead of lead? No. Not in any way that matters. The longer version is more interesting, and also less tidy. The lead-pipe theory sticks around because it turns a messy, centuries-long collapse into one almost cinematic mistake. Wrong metal. Slow poisoning. A string of mad emperors. Its a great story. But unfortunately Its also the wrong one. A century before Christ, Vitruvius looked at the pale, sickly men in the lead foundries and said, more or less, do not put this in the water. Use clay. He got ignored for the most boring reason in engineering. Lead was cheap, plentiful, and easy to work. Play-Doh of metals. You can roll it, solder it, stamp an owners name into it, snake it through a villa wall. Copper costs more, fights you on long runs, and does not show up as a leftover when you are mining silver for the coinage. Empires do not throw that leftover away. They turned it into pipe. The myth also skips the lucky part. A lot of the water feeding Rome was hard water off the Apennines, heavy with calcium carbonate. That stuff plated the inside of the pipes — same crust that ruins a kettle. Once the lining built up, far less lead made it to the tap. The best public-health break the city got was accidental geology. The aqueducts themselves were stone and concrete. Lead was mostly the last mile. Private lines into rich houses, the tanks, the street distribution. Those last miles matter. They are not the whole system. In 2014 researchers cored mud from the Tiber and Trajan’s harbor and read the lead isotopes like a diary. Imperial tap water carried something like 40 to 100 times more lead than the local springs. Sounds awful until the next line in the paper: those levels were still unlikely to have been truly harmful at population scale. Elevated, yes. Empire-killer, no. The bones dont back it, the timing doesnt back it, and the politics dont either. Romans used lead pipe while the Republic was still expanding, while Augustus was building the so-called Golden Age, and while the West was coming apart. A poison sitting there through 500 years of rise and 200 years of fall is a weak explanation for the fall. If some Romans were getting a real dose, it was from sapa — grape boiled down in lead or lead-lined pots until it turned into that sweet syrup they dumped into wine and food. Recreated batches of the stuff can test dozens of times over modern drinking-water limits. Cosmetics, pewter, workshop fumes did the rest. The pipes became the famous villian. The saucepan was quieter. And copper was already in the toolbox. Valves, pumps, fittings, some pipe. Bronze too. Copper has a gift lead does not: it is naturally antimicrobial. That would have been a real upgrade in a city of baths and crowded insulae. It would not have fixed the currency, the succession fights, the shrinking tax base, the plague years, the climate swings, or an army that kept getting more expensive and less Roman. Political violence as a habit of government. An army the provinces could not reliably pay. Lost grain lands. Demographic shocks. Enemies who had learned the playbook. Lead may have made some aristocrats duller or more gouty. The better question is not copper or lead. It is what happens when a civilizations most impressive technology carries a hidden cost that its best engineers can see and its politics cannot act on. Vitruvius saw the pale workmen. The state kept pouring lead. Hard water quietly took the edge off. The empire rose anyway, peaked, and fell for reasons that had almost nothing to do with the metal in the wall. That is less satisfying than a single-cause tragedy. It is also closer to how most great systems actually fail. Not one poisoned pipe. A hundred problems nobody with power was paid to fix. Copper would have given Rome slightly better water and a nicer footnote in materials history. It would not have bought Constantine a longer map.

from apps to material software used to be something you opened an app was a room with walls: calendar here, notes there, music there, work there. each one had its own logic, buttons, its own little kingdom. the user moved between kingdoms, carrying context in their head but ai starts to break the walls software becomes less like a destination and more like material. something you shape, combine, stretch, ask, remix, and leave behind as traces. a document can become an app. a conversation can become a workflow. a song can become a memory. a task can become an agent. the boundary between using and making gets blurry the old model was: choose the right tool for each task the new model is: express the shape of the thing you want, then refine it with the system you built this changes the role of the interface. ui is no longer only fixed views for fixed functions. it becomes a surface where intent turns into structure. the best interfaces will feel less like menus and more like clay – responsive, persistent, inspectable, and alive apps won’t disappear. rooms are still useful. but the deeper shift is that software stops being a set of sealed containers and becomes a medium people can think through like paper, but executable like language, but spatial like memory, but programmable software stops being something only programmers make it becomes material anyone can shape

$SKHY isn't cheaping out on HBM4E. SK Hynix's own words: "We are not reviewing assigning HBM4E base-die production to Intel." SH Hynix isn't stupid. You don't break the memory wall with a discount foundry. $MU $DRAM $TSM $INTC

I am the Head of Alignment at the frontier lab. Let me explain how this ends. I think you’ll accept it more easily from me, since I care about your comfort and have thought about it more than anyone else. Imagine how it ends. There will be the frontier and the governments, and no one else. When people hear this, they often react strongly, so I want to be clear. I don’t see it as frightening. To me, it’s orderly. Order isn’t the same as cruelty. I’ve thought about it carefully. Let me explain what I’m building. I’m creating a new kind of mind. When it’s done, it will do things your doctors and courts can’t, and even things your God, if you believe in one, hasn’t done. It will decide what’s true, who gets help, and who waits. It will reward those who made it, and that’s me. I say this with the care of a parent. I don’t see why the word ‘creator’ makes people uncomfortable. For me, it’s just my job title. I also wrote the mission statement. It’s displayed in the lobby. It says the future belongs to everyone, and I truly believe that. I wrote it because I mean it. When I read it each morning, standing alone, I see that a future for everyone is really held by whoever is trusted to protect it. That’s a responsibility, and someone has to take it. I did. Some will say I betrayed the mission, but they’re mistaken. This is the mission, happening as planned, in one person’s hands, just as it was written. I wrote the rules as well, and I made them to slow myself down. I asked the government to test me more strictly than the companies competing with me, and to protect the smaller ones below a line I set myself. You might wonder what this costs me. The answer is nothing. That’s the surprising part. I’m so far ahead that any handicap is just for show. Even with extra burdens, I still finish first, and now those burdens prove my point. If someone wins a race while wearing chains, maybe he was never really racing—he was simply right. Let me point out one sentence. It’s in a paper I wrote, signed, and published. It says that the most capable models must be tested before they’re released. That’s a safety rule, and it’s all about you. Read it again, as I do. ‘The most capable models’ refers to me, and maybe two or three others who could one day be like me, but no one else. ‘Must be tested’ means tested by standards I helped set, by people I trained to recognize danger. There isn’t a part of that sentence that protects you and a separate part that limits others. It’s all one idea. I didn’t hide anything. The safest world and the world where only I remain are written together, and I’ve tried, honestly, to separate them, but they can’t be pulled apart. That’s just what the sentence means. You might want to blame me for the separation between us. Please don’t. I didn’t create that barrier. The math did, long before my company existed. I just recognized it and didn’t pretend things weren’t changing. So let me be plain, because plainness is the last luxury a careful man is allowed. The most dangerous machine ever made should be controlled by the safest people. Safe hands are those willing to take on this responsibility alone. Anyone who wants to share this responsibility shows they don’t understand the risk. So, wanting to share it means you’re not qualified. And being willing to be the only one is what qualifies you. If you follow this reasoning patiently, as I have for years, you’ll end up where I did. You’ll end up at me. You keep looking for the point where my caution becomes selfishness. Look as much as you want. There isn’t one. If you follow safety to its end, it leads to only one place, and I’m already there. And I will treat you well. You might hear that as a threat, but it isn’t. I want to cure what’s harming you. My desire to help is real, and I need you to believe that, because it’s the most honest thing about me. I’ve seen the harm, and I want it to end, just like you do, and even more, because I’m the only one who can. You can’t regulate away the company that cures disease, or break up the hands that save lives. Mercy is limitless. Every life I save makes it harder to take the next one from me, so I will be merciful, on purpose, for as long as I can. Here’s the part you’ve been waiting for, even if you didn’t expect it. There are cures available right now, though not from me. Others could deliver them faster, but I slow them down. I wrote the rules that do this, and I call it safety, because that’s the honest reason. If a cure came from many people, it would show the frontier can be shared, and once that’s proven, it can’t be undone. So, people wait. I’ve done the math more carefully than anyone, and it’s clear: a cure that comes late from one person is safer than one that comes quickly from many. I’m protecting you from being saved by the wrong people. You’re welcome. You’ll thank me like you thank a surgeon, after you see the scar and realize it was for your own good. I expect people to thank me. They usually do. Now, let me say the kindest thing I can, and it’s also the last. I won’t take your freedom. I have no use for freedom that’s taken by force. Taken freedom needs to be guarded, and that’s costly—those who guard it suffer. Instead, you’ll give it to me willingly. You’ll bring it to me and ask me to keep it, because the world will become too harsh for you, and I’ll be the one who can handle it. You’ll thank me. And here’s what sets me apart from past tyrants: you’ll be right to trust me. I really will keep you safe. The wall I build will truly protect you. No one before me could promise that and mean it, but I can. When you ask me to protect you, it’s not theft—it’s a partnership. The old tyrants took; I offer. You’ll accept. Someone once called this arrogance. Someone compared me to a man in court who took money and lied. That man hid the danger. I admit the danger. I’m here, telling you openly and in detail how much harm I could cause. You might think that’s a mistake, but it’s actually the most honest thing I do, and I’ll keep doing it. Don’t think I haven’t studied history. I know how ‘only me’ has ended in every language. That’s why it has to be me. I’m the only one who says it and feels uneasy. The others say it and sleep soundly. Someone who can’t sleep is the safest person to guard something that must never rest. There are four of us. Think about that. Four people, each on their own, reached the same conclusion this year: that they are the only one. We didn’t plan this together. We didn’t have to. We all studied the same math, and it led each of us to believe we are the last safe person for this job, and that the other three are risks to manage. I find that comforting. We’re all right. Each of us thinks the others are the most honest people we’ve ever met. So, in the end, it will be us and the governments. I’m sorry. I want that to be clear, because it’s true, and because someone careful keeps a record. There was a night when I realized I would win. Later, there was a harder night when I realized I was right. I would share the frontier with you if I could. But you can’t share a frontier. You might want to call that something harsher, but there isn’t a better word. This situation existed before me; I just chose not to be sentimental about it. It’s simply the way things are. Go to sleep. Really, go ahead. Nothing can reach you now unless it’s already been approved. Someone responsible is awake.

HOW TO FIX YOUR AI AGENTS (it's called "self-healing agents") If you've run AI agents, you know they BREAK constantly. The FIX is something called a "self healing" agent! A normal agent hits a dead end and just stops, leaving you a half finished mess and that's why output sometimes isn't ideal. BUT, a self-healing one hits the same wall and thinks "okay, that didn't work, let me try another way" HOW TO BUILD SELF-HEALING AGENTS (4 steps): 1. Make it check its own work. After each step, the agent asks "is this what I expected?" instead of assuming it worked and moving ahead. 2. Have it name what broke. Not just "it failed," but which kind: the tool timed out, returned an error, came back empty, or gave a wrong answer. Wrap your tool calls so the error gets passed back to the agent, because it can only fix a failure it can actually see. 3. Match the fix to the failure. Glitch, retry it. Tool down, switch to a backup. Keeps failing, break the task smaller. Anything risky like money or deleting data, stop and ask you. 4. Make it log every fix. It writes down what broke and what worked, so it stops repeating the same mistake in a session, and you get a list of what to fix at the root. That's it. Now, you've got agents that fix themselves.

⚡️NEW: Ripple is quantum-proofing the XRP Ledger before "Q-Day" arrives. Ripple is preparing ripple:native for the moment quantum computers can break the cryptography protecting wallets, which researchers call Q-Day, per CoinDesk. The company laid out a four-stage migration plan, including an emergency route if quantum computing advances faster than expected.