From Hallucinations to Clicks Re-claiming hallucinated 404 traffic on websites with many thousands or millions of URLs. dejan.ai/blog/from-hallucina…

I hate how people use the word “aesthetic” these days. Saying something “is aesthetic” is retarded & wrong. Something can be “aesthetically pleasing” or something may have “an appealing aesthetic”

In case you missed it: Gemini (and consequently AI Mode) IS BEING TRAINED by Google's traditional search signals (cough, *PageRank*) Talk about the death of link building or SEO 😎 /weekend reading

Which tech couple are you betting on?

i OnLy UsE yOu To DoWnLoAd ChRoMe

I simply do not believe you.

I got banned from AI Mode for doing this, but now we know what we didn't know: dejan.ai/blog/ai-mode-intern…

Google’s AI Mode is all about follow-up questions. I built a free tool to simulate that logic: → detects core entities → generates follow-ups → stress-tests your content Built with @DSPyOSS. No CTR, no tracking—just structure intent. 🔗 wordlift.io/blog/en/query-fa… #SEO

Gridstack Blocks are nearly finished. Toolbar for editing, position saves, posts as tiles, and reuse tiles across many grids.

waking up and realizing theres no longer a single thing im better at than AI is, given the right context

Programmers suck at API design. Never in my entire life have I ever wanted any createFile operation to fail because the folder didn’t exist. I said make file. MAKE THE FILE. Make directories if needed. This isn’t effing hard. Everyone is fired into the sun.

sitting in the press room here at #googleio and thinking, how many of these folks will be out of a job soon because of Google

Google Search "Discussions" beta feature may one day replace Reddit results in Google... Maybe... seroundtable.com/google-sear… via @Callmenicholi and @glenngabe

We're missing (at least one) major paradigm for LLM learning. Not sure what to call it, possibly it has a name - system prompt learning? Pretraining is for knowledge. Finetuning (SL/RL) is for habitual behavior. Both of these involve a change in parameters but a lot of human learning feels more like a change in system prompt. You encounter a problem, figure something out, then "remember" something in fairly explicit terms for the next time. E.g. "It seems when I encounter this and that kind of a problem, I should try this and that kind of an approach/solution". It feels more like taking notes for yourself, i.e. something like the "Memory" feature but not to store per-user random facts, but general/global problem solving knowledge and strategies. LLMs are quite literally like the guy in Memento, except we haven't given them their scratchpad yet. Note that this paradigm is also significantly more powerful and data efficient because a knowledge-guided "review" stage is a significantly higher dimensional feedback channel than a reward scaler. I was prompted to jot down this shower of thoughts after reading through Claude's system prompt, which currently seems to be around 17,000 words, specifying not just basic behavior style/preferences (e.g. refuse various requests related to song lyrics) but also a large amount of general problem solving strategies, e.g.: "If Claude is asked to count words, letters, and characters, it thinks step by step before answering the person. It explicitly counts the words, letters, or characters by assigning a number to each. It only answers the person once it has performed this explicit counting step." This is to help Claude solve 'r' in strawberry etc. Imo this is not the kind of problem solving knowledge that should be baked into weights via Reinforcement Learning, or least not immediately/exclusively. And it certainly shouldn't come from human engineers writing system prompts by hand. It should come from System Prompt learning, which resembles RL in the setup, with the exception of the learning algorithm (edits vs gradient descent). A large section of the LLM system prompt could be written via system prompt learning, it would look a bit like the LLM writing a book for itself on how to solve problems. If this works it would be a new/powerful learning paradigm. With a lot of details left to figure out (how do the edits work? can/should you learn the edit system? how do you gradually move knowledge from the explicit system text to habitual weights, as humans seem to do? etc.).

what is the greatest art produced by millennials? this question baffles me

Introducing the Orb Mini. Designed to make human verification more accessible, it’s transportable and seamlessly accommodates your lifestyle. The Orb Mini, it goes where you go.