atomless

In late February, the British singer-songwriter Benedict Cork posted a snippet of himself playing a soulful new song. A few days later, Cork started receiving messages about the song’s release—which was odd, because he hadn’t finished it yet. Eventually, someone sent him the song, which was on streaming platforms under someone else's name. This version of his song had been created with AI. Someone had seen Cork’s snippet, and, trying to capitalize on its virality, ran it through an AI music generator and posted the result in order to generate streams. Cork is one of a growing number of musicians facing increased competition or outright theft from AI music generators. Some musicians contend that AI is helping to democratize the creation of music. But many musicians fear the entire industry is sliding towards prioritizing machine-made slop, making it increasingly hard for rising or independent musicians to make a living. Tech correspondent Andrew Chow explains the growing world of AI music slop. Read more here: time-magazine.visitlink.me/jvIAnR

I still can’t get my head wrapped around this one. The president is having a satellite burnt up because he doesn’t like the information It is sending back to us telling us we are killing the planet.

Eric Hobsbawm defined darkness not as a apocalyptic break but as the state where everyone "gets used to living under conditions which should not be tolerated". Getting used to livestream genocide, carpet bombings and now threat of imminent nuclear attack is living in darkness.

This is beyond shameful: x.com/atrupar/status… When asked if it "bothered him at all" when Trump "threatened to kill the entire Iranian civilization," Rutte's response was simply: "what I want you to know is I support the President and I know a large part of Europe do." If Europe cared at all about the "values" it supposedly has, this should be cause for immediate dismissal, and potentially criminal prosecution - publicly endorsing genocide would be classified as illegal hate speech in several EU countries, including in the Netherlands (Rutte's own country).

On the first day of the ceasefire, Israel launched its most aggressive attack on Lebanon. Why, if not to cancel the ceasefire?

From a friend in Beirut: “They bombed highly populated and non-evacuated areas. Last I heard there were 254 dead and 1169 injured. The day before, at the time they bombed, (my husband) and I were passing in that area. The casualties will rise because there are still people under the rubble. This is higher than the Beirut port explosion. And they are threatening that they will continue their unannounced raids for 48 hours. So nobody knows where they are going to bomb today. We don’t know anymore where it’s safe. And the sad thing is that they are getting away with everything they do and no one is able to stop them.” All this because Benjamin Netanyahu understands that peace will mean the end of his tenure as Israel’s Prime Minister. The people of Lebanon suffer because of the ego of one man. A man who heads a nation of genocidal maniacs. Where is Europe? Where is Russia? Where is China? Don’t bother asking where America is. We lost our moral compass a long time ago. We stand with Israel. We stand with the murderers.

So-called "AI" Companies are addicting people to magic beans. Bought by the packet, these are legumes harvested from the soils of stolen & poverty waged labour. Each bean can be exchanged only with the Co they were bought from, and for an amount of beanstalk the Co keeps secret.

Stewart Lee is bang on here with this analysis. Comedy won't tell the truth, the question is why? follow the money. The answer lies in the production companies, who owns them? via @reginalddhunter

A woman trapped in a damaged building waits to be rescued following an Israeli strike, in Beirut, Lebanon, April 8, 2026. REUTERS/Emilie Madi

While massive crowds are gathering at our campaign events, the regime, gripped by fear, has shifted into panic mode. They have fired up the shredders, and the disintegration of the power structure has begun. In exactly four days, the decisive parliamentary election will conclude. Now is the time to push forward, now is the time to mobilize all our energy. Under a TISZA government, order, peace, and security will return to Hungary. And, of course, there will be more smiles and more joy. Long live a free, independent, European, and humane Hungary.❤️🤍💚

I don’t think people around the world realize how bad the attack on Lebanon was today. It’s BAD. Like 9/11 BAD. Hospitals are overflowing and do not have enough blood. We still don’t know what the death toll is. This is TERRORISM. Pure and Simple. April 8. Never forget.

The most advanced ai on earth

🚨 CONCERNING: Stanford just published a paper that should alarm every company building multi-agent AI. When thinking tokens are matched, single agents beat debate systems, parallel role systems, ensemble agents, and sequential pipelines. The multi-agent advantage is a compute accounting artifact not an architectural breakthrough. Stanford tested single agents against five different multi-agent architectures across three model families Qwen3, DeepSeek-R1, and Gemini 2.5 on multi-hop reasoning tasks. The key variable: thinking tokens held constant across every comparison. When compute is equal, single agents match or outperform every multi-agent design tested. Every time. The reason is mathematical, not empirical. Multi-agent systems pass information between agents as messages. Every message is a compressed, lossy version of the full context. The Data Processing Inequality proves that no downstream agent can recover information discarded in that compression. A single agent with access to the full context is information-theoretically guaranteed to perform at least as well as any multi-agent system operating on summaries of that context. Stanford then ran the numbers. Results across all models and budgets: → Single agent average accuracy at 1000 tokens: 0.418 → Sequential pipeline: 0.379 → Subtask-parallel: 0.369 → Parallel roles: 0.381 → Debate: 0.388 → Ensemble: 0.333 Not one multi-agent architecture beat the single agent at any matched budget above 100 tokens. The pattern held across Qwen3, DeepSeek, Gemini 2.5 Flash, and Gemini 2.5 Pro. It held across two different benchmarks. It held across six different token budgets from 100 to 10,000. Stanford also found a significant measurement artifact in the Gemini API. When you request 10,000 thinking tokens, the API reports 1,687 tokens used. The visible thought text contains an average of 251 words — roughly 359 tokens. That's a 4.7x inflation factor. Multi-agent systems produce more visible thought text than single agents under the same requested budget because multiple agent calls generate multiple thought blocks. This makes multi-agent systems look like they're reasoning more when they're just generating more text. Every benchmark that didn't control for this is measuring compute, not architecture. There is one regime where multi-agent systems become competitive: corrupted context. When 70% of the reasoning context is replaced with random tokens, sequential pipelines start outperforming single agents. When misleading information is injected into the context, multi-agent decomposition helps filter it. But under normal conditions with clean context and matched compute — single agents win. Most reported multi-agent gains come from one of two sources: → Unaccounted compute multi-agent systems simply use more tokens → Context degradation single agents struggle when context is noisy or corrupted Neither is an architectural advantage. Neither justifies the complexity. The question every AI team should ask before building a multi-agent pipeline: Are you controlling for thinking tokens? If not, you're not measuring whether your architecture works. You're measuring whether more compute helps. It always does.

Israel ruthlessly carpet bombed Lebanon today, killing at least 180 people and injuring more than 800. How does the BBC report it? “Israel says it hit more than 100 command centres and military sites in 10 minutes…” This is how war crimes are laundered.

A Trump insider opened a $51,000,000 oil short position — hours before Trump announced a ceasefire with Iran. This guy is now 16 for 16. $170 million in profit. A perfect streak. This is not a talented trader. "We placed the bet." "The ceasefire dropped." "We cashed out." Sixteen times in a row. That is not skill. That is not instinct. That is not research. That is someone who knows what is coming before it comes. Think about what that actually means. A private individual is placing a $51 million bet that oil prices are about to collapse — hours before a sitting president announces a ceasefire that collapses oil prices. Not once. Sixteen times. Zero losses. There are only two explanations and both should terrify you. Either someone inside the White House — or with direct access to it — is leaking ceasefire negotiations to traders before diplomats, before the press, before the American people hear a single word. That is insider trading. That is corruption. That is a federal crime. Or the timing of the announcement itself is being shaped around the trade. Which is worse. This is not a genius investor who reads the news faster than you do. The news hadn't happened yet. He wasn't reading the news. He was getting a phone call. While Americans were watching the ceasefire announcement and feeling relieved — somebody already knew. Somebody had already bet $51 million on it. And somebody was already counting their winnings. You are not watching a free market. You are watching a White House with a side hustle. Via~ Really American

It was always clear that Gaza would not and could not be contained. It is why we are here, on the precipice of calamity in Iran.

Did anyone expect Netanyahu's Israel to honour a ceasfire?

Israel is the most dangerous, destructive, destabilising force in the region.

loved the new academic book "Techno-Negative" by Thomas Dekeyser, a deep history of resisting and destroying technology, way beyond Luddites. I reviewed it for @NewYorker. recommended reading for tech industry & writers

The ICE warehouse spending spree may be padding the accounts of big financial institutions, all at your expense. New data shows DHS paying investment firm Blue Owl Capital $119 million for a warehouse, 2x the market value. And at least 33 members of the Trump administration invest in Blue Owl’s various private equity funds. In another case, Goldman Sachs sold a New Jersey facility to ICE for $129.3 million — 137% over value. And that's just the tip of the iceberg. Read the full story here: substack.perfectunion.us/p/the-worlds-b….