@_deepfates went in sermon mode this week and I love it: https://t.co/p7SQdV9YnB
In three words: deep learning worked.
In 15 words: deep learning worked, got predictably better with scale, and we dedicated increasing resources to it.
That’s really it; humanity discovered an algorithm that could really, truly learn any distribution of data (or really, the underlying “rules” that produce any distribution of data). To a shocking deg... See more
In 15 words: deep learning worked, got predictably better with scale, and we dedicated increasing resources to it.
That’s really it; humanity discovered an algorithm that could really, truly learn any distribution of data (or really, the underlying “rules” that produce any distribution of data). To a shocking deg... See more
samaltman.com • The Intelligence Age
Another nuance is that most companies and developers do actually not self-host open source LLMs, but use services like together.ai to consume them. Thus, practically speaking, most developers consume open source models in the same fashion as they consume OpenAI or Anthropic models - via third party APIs. Also, most large models like DeepSeek R1 are... See more
“There’s this great essay by Hannah Baer in the latest issue of Artforum, where she unpacks why we are so afraid of AI becoming more intelligent than humans. If you look at this historically, humans have positioned themselves as the most intelligent species, even though that’s not true. As the “most intelligent species,” we’ve used this power as an... See more
Meet Mindy Seu — passerby magazine

And in the end, OpenAI doesn’t matter . They are making the same mistakes we are in their posture relative to open source, and their ability to maintain an edge is necessarily in question. Open source alternatives can and will eventually eclipse them unless they change their stance. In this respect, at least, we can make the first move.
Simon Willison • Leaked Google document: “We Have No Moat, And Neither Does OpenAI”
The rise of AI is perhaps best understood as a working-class positive singularity, as centuries of accumulated technocratic knowledge, in one fell swoop, is gifted to the working classes. It might be compared to the introduction of longbows in our reference period. The models that have been open-sourced are already good enough to permanently weaken... See more