Sen LinYu

he rearrangement of social and economic conditions around the pursuit of attention is a transformation as profound as the dawn of industrial capitalism and the creation of wage labor as the central form of human toil. Attention now exists as a commodity in the same way labor did in the early years of industrial capitalism. What had previously been... See more

. There is a quiet allure to someone who is absorbed in their own interests—they feel layered, self-contained, and slightly undiscovered

Bourdain taught me that culture is more than aesthetics. It’s how people survive. It’s what they sing when they’re hurting. It’s the rituals of joy in the middle of chaos. It’s rock shows and battered notebooks, food stalls and high art, leather jackets and hand-me-down wisdom.

State investment into public education and cultural infrastructure gave the UK some of its most important contemporary cultural movements: kitchen-sink realism, post-punk, Britpop - waves of creativity built by working-class kids who suddenly had access to literacy, art, theory, rehearsal rooms, grants, time. But as higher education is economically... See more

And of course, unlike what you wear, knowledge lives on the inside. So this internal accumulation has to be performed externally through taste, references, opinions. It echoes the boom of platforms like Goodreads or Letterboxd. We no longer enjoy knowledge silently. In a wider culture where value is tied to metrics and monetisation, knowledge that... See more

Those capable of analytical thought outside algorithmic recommendation gain disproportionate power, while the rest sink deeper into passively generated preferences.

In June, researchers at OpenAI reported the results of their own tests of emergent misalignment (opens a new tab). Their work suggests that during pretraining, an AI learns a variety of personality types, which the researchers call personas. Fine-tuning the model on insecure code or incorrect medical advice can amplify a “misaligned persona” — one... See more

Many AI researchers use the word “emergence” to describe behaviors or actions that a model can exhibit for which it wasn’t trained. In the last few years, myriad experiments have shown, for example, that large language models, trained only on text, can produce emergent behaviors like solving simple arithmetic problems or generating computer code.

noticing when you’re pattern-matching someone into a category instead of actually hearing what they’re saying. Or catching yourself mid-assumption and asking a question instead.