AI - Ongoing Model Improvements

t doesn’t help that the two most impressive implementations of AI for real work - Claude’s artifacts and ChatGPT’s Code Interpreter - are often hidden and opaque

The incentive to find breakthrough science that provides a performance pathway other than scaling is increasing. A GPT-6 class model will cost $10 billion and three years to improve on GPT-5 by some uncertain degree. That’s a ton of time and a lot of cash for an uncertain payout: in other words, a substantial prize for anyone who can figure out pro... See more

As Ethan said a year ago, it takes 10 hours of monkeying around to get into it. But I think a good AI executive training program can dramatically shorten the window, assuming a growth mindset on the exec’s part. I ran into a retired architect the other day who was bemoaning wanting to work in retirement but he had never learned CAD-CAM! Now he’s tw... See more

First, operationalisation takes a long time, often revealing hidden potential in existing technologies. As Jack Clark notes, “if we stopped all AI progress today, there’s a huge capability overhang”. Even without further model development, building the right scaffolding can unlock surprising capabilities. This scaffolding isn’t just software; it in... See more

While LLMs continue to devour web-scraped data, they’ll increasingly consume their own digital progeny as AI-generated content continues to flood the internet. This recursive loop, experimentally confirmed, erodes the true data landscape. Rare events vanish first. Models churn out likely sequences from the original pool while injecting their own un... See more

Arbituram

Jul 4

Same here; I'm not rubbishing the benchmarks, they have their uses, but Sonnet 3.5 just *feels* so far ahead of anything else, I had to go for a long walk on the beach after using it to contemplate how long humanity has left.

Sonnet 3.5 is able to have thoughtful and nuanced conversations on complex topics and to be proactive in codin... See more

it is kind of surprising that none of the major AI labs seem to have put out any deep documentation aimed at non-specialists. There are some guides for programmers or serious prompt engineers, but remarkably little aimed at non-technical folks who actually want to use these systems to do stuff - the vast majority of users

But there's just something about the immediacy and interactivity of Claude's "Artifacts" window, combined with the model's less stilted tone that brings it home.

For the most avarege users, these seemingly minor difference matter so much more than raw performance on abstract LLM benchmarks.

LeCun points to four essential characteristics of human intelligence that current AI systems, including LLMs, can’t replicate: reasoning, planning, persistent memory, and understanding the physical world. He stresses that LLMs’ reliance on textual data severely limits their understanding of reality: “We’re easily fooled into thinking they are intel... See more