Join 📚 J'S Highlights

A batch of the best highlights from what J's read, .

With online clout too often determining movement leadership, and few ways to hold those leaders accountable, conflict and mistrust easily spread, supercharged by algorithms hooked on rage and egged on by fake accounts and Russian bots that heaped salt on all our open wounds. The predatory corporate logics that earlier iterations of the left recognized as our enemies were clearly deep inside us now: in the arteries that connected us, in our habits of mind, in our very cells. It felt like nothing could be trusted, least of all one another.

Doppelganger

Naomi Klein

You can spin up your own hosted GPT4All on @modal_labs in 10 lines of code! https://t.co/VmxtynXSId https://t.co/P7wmf4F0kk ![](https://pbs.twimg.com/media/FwK_zt4WAAQjHDe.jpg)

Tweets From AndriyMulyar

@andriy_mulyar on Twitter

As predictive text-generators, Generative AI models do not assess whether the outputs they generate are true or not. Rather, they produce streams of text based on, albeit sophisticated, statistical probability. Thus, their outputs have been compared to Frankfurtian bullshitters—those who are indifferent to the truth (or falsity) of their statements.

2508.19588

Lucy Osler

...catch up on these, and many more highlights