The second screenshot goes to a chart where the Y axis is labelled
Task duration (for humans) where logistic regression of our data predicts the AI has a 50% chance of succeeding
So they're just extrapolating an exponential, not actually measuring it.
Check out this epic cope from an Anthropic employee desperately trying to convince himself and others that actually LLMs are getting exponentially better
https://www.julian.ac/blog/2025/09/27/failing-to-understand-the-exponential-again/
Includes screenshots of data where he really really hopes you don't look at the source, and links to AI 2027.
He's been on the vibecoding bandwagon for quite some time on lobste.rs.
What book is that?
Not sure if a LessWronger extensively quoting Accelerando as a blueprint for a singularity takeoff is complimentary to Charles Stross:
Worht noting is that the novel is 20 years old...
The UK public being skeptical about AI must be good news for Big Yud.
Usual suspects weighing in.
Remind me: why is Hyprland bad? I don't doubt it is, I just need it to add to my mental list of stuff to avoid.
Harvard Business Review: AI-Generated “Workslop” Is Destroying Productivity
[...] Employees are using AI tools to create low-effort, passable looking work that ends up creating more work for their coworkers. On social media, which is increasingly clogged with low-quality AI-generated posts, this content is often referred to as “AI slop.” In the context of work, we refer to this phenomenon as “workslop.” We define workslop as AI generated work content that masquerades as good work, but lacks the substance to meaningfully advance a given task.
Cloudflare sponsors Ladybird and Omarchy, techfash workfare
They say "Luddites" as if it's a bad thing...
Surprise level: zero