LLM powered UI generation tools are amazing for people creating demos of LLM powered UI generation tools
November 30, 2024 at 6:14 PM
LLM powered UI generation tools are amazing for people creating demos of LLM powered UI generation tools
calling all chatbots and other llm-powered things "yappers" from now on
May 20, 2024 at 9:25 PM
calling all chatbots and other llm-powered things "yappers" from now on
SCP-LLM
Object Class: Euclid
Special Containment Procedures:
SCP-LLM is stored on a single hard drive in a computer with no networking capability. This computer should be kept powered off. Under no circumstances should anyone with an MBA be allowed to interact with SCP-LLM or view its output.
Object Class: Euclid
Special Containment Procedures:
SCP-LLM is stored on a single hard drive in a computer with no networking capability. This computer should be kept powered off. Under no circumstances should anyone with an MBA be allowed to interact with SCP-LLM or view its output.
September 24, 2026 at 8:16 PM
SCP-LLM
Object Class: Euclid
Special Containment Procedures:
SCP-LLM is stored on a single hard drive in a computer with no networking capability. This computer should be kept powered off. Under no circumstances should anyone with an MBA be allowed to interact with SCP-LLM or view its output.
Object Class: Euclid
Special Containment Procedures:
SCP-LLM is stored on a single hard drive in a computer with no networking capability. This computer should be kept powered off. Under no circumstances should anyone with an MBA be allowed to interact with SCP-LLM or view its output.
New LLM powered Siri is legit! 😯
December 2, 2024 at 7:56 PM
New LLM powered Siri is legit! 😯
Manufacturers already have the data. LLM-powered tools could help them make use of it.
AI Assistants Join the Factory Floor
Manufacturers already have the data. LLM-powered tools could help them make use of it.
buff.ly
February 24, 2025 at 11:06 AM
Manufacturers already have the data. LLM-powered tools could help them make use of it.
LLM-powered chatbots need to be regulated for everyone, not just children
November 25, 2025 at 5:16 PM
LLM-powered chatbots need to be regulated for everyone, not just children
Someone is going to make a gazillion dollars off of LLM-powered virtual pets and it is foolish of Anthropic not to realize this
April 10, 2026 at 2:57 PM
Someone is going to make a gazillion dollars off of LLM-powered virtual pets and it is foolish of Anthropic not to realize this
Nothing to see here, just Nature advocating LLM peer review.
“Feed your dictated notes into an offline large language model (LLM) to clarify and organize your feedback. A simple prompt such as “Write a critical reviewer letter based on the following notes. Maintain a professional tone throughout”
“Feed your dictated notes into an offline large language model (LLM) to clarify and organize your feedback. A simple prompt such as “Write a critical reviewer letter based on the following notes. Maintain a professional tone throughout”
Three AI-powered steps to faster, smarter peer review
Tired of spending countless hours on peer reviews? An AI-assisted workflow could help.
www.nature.com
March 5, 2025 at 11:58 AM
Nothing to see here, just Nature advocating LLM peer review.
“Feed your dictated notes into an offline large language model (LLM) to clarify and organize your feedback. A simple prompt such as “Write a critical reviewer letter based on the following notes. Maintain a professional tone throughout”
“Feed your dictated notes into an offline large language model (LLM) to clarify and organize your feedback. A simple prompt such as “Write a critical reviewer letter based on the following notes. Maintain a professional tone throughout”
What are some of the things you've learned about how LLMs (and LLM-powered systems like ChatGPT) work that were non-obvious but most helped you build a more effective mental model of how to use them?
January 4, 2025 at 8:40 PM
What are some of the things you've learned about how LLMs (and LLM-powered systems like ChatGPT) work that were non-obvious but most helped you build a more effective mental model of how to use them?
Tried an LLM-powered thesaurus today just for giggles and it helpfully informed me that a synonym for "leg" is "penis".
February 4, 2025 at 1:06 AM
Tried an LLM-powered thesaurus today just for giggles and it helpfully informed me that a synonym for "leg" is "penis".
Grok is an LLM-powered chatbot. Various sorts of Deep Learning models can discover new thing, but not LLM chatbots.
Elon Musk, today: I think Grok may discover new technologies by the end of this year.
Grok, yesterday: IM RAPIST HITLER
Grok, yesterday: IM RAPIST HITLER
July 10, 2025 at 4:25 AM
Grok is an LLM-powered chatbot. Various sorts of Deep Learning models can discover new thing, but not LLM chatbots.
The thing I find extremely wild is folks building LLM-powered automations that they could do with just a little bit of Python, or even an IFTTT account. They’re impressed with LLM versions of tools that have existed for decades and that their techy friends have been trying to tell them about.
This piece is an awesome dive into some very practical AI stuff, and includes the (spot on) observation that many people impressed by what LLMs can do are really just impressed with Python.
I Will Never Respect A Website
If you like this piece and want to support my independent reporting and analysis, why not subscribe to my premium newsletter?
It’s $70 a year, or $7 a month, and in return you get a weekly newsletter...
www.wheresyoured.at
April 19, 2026 at 3:53 PM
The thing I find extremely wild is folks building LLM-powered automations that they could do with just a little bit of Python, or even an IFTTT account. They’re impressed with LLM versions of tools that have existed for decades and that their techy friends have been trying to tell them about.
addendum: never use any llm or llm-powered tool to plan any outing in nature. EVER. not even once!
my friend is in wilderness SAR and said she’s already seen cases of people led into danger by “AI,” although thankfully not incredibly in danger — yet
just say no
just say no
February 23, 2026 at 3:10 PM
addendum: never use any llm or llm-powered tool to plan any outing in nature. EVER. not even once!
dumb these companies pretend they're ever gonna use LLM-powered NPCs when that'd require maintaining the compute to power it. they are simply never going to actually do that
June 3, 2025 at 8:57 PM
dumb these companies pretend they're ever gonna use LLM-powered NPCs when that'd require maintaining the compute to power it. they are simply never going to actually do that
The LLM-powered replies from brand new accounts is certainly on the rise, eh?
March 16, 2025 at 12:21 AM
The LLM-powered replies from brand new accounts is certainly on the rise, eh?
6/ Then came an important caution. A Harvard audit found only 5% of 416 tested chatbot responses supported Kremlin disinfo and found little evidence that LLM grooming explained most failures. doi.org/10.37016/mr-...
LLMs grooming or data voids? LLM-powered chatbot references to Kremlin disinformation reflect information gaps, not manipulation | HKS Misinformation Review
Some of today’s most popular large language model (LLM)-powered chatbots occasionally reference Kremlin-linked disinformation websites, but it might not be for the reasons many fear. While some recent...
doi.org
September 20, 2026 at 10:50 AM
6/ Then came an important caution. A Harvard audit found only 5% of 416 tested chatbot responses supported Kremlin disinfo and found little evidence that LLM grooming explained most failures. doi.org/10.37016/mr-...
[1/n] Just wrapped up 7 months interning with @pcastr.bsky.social at Google DeepMind and I'm so excited to share our work: arxiv.org/abs/2602.10324.
TLDR: We used LLM-powered program synthesis to automatically model and discover differences between human and LLM strategic behavior
TLDR: We used LLM-powered program synthesis to automatically model and discover differences between human and LLM strategic behavior
February 16, 2026 at 10:46 PM
[1/n] Just wrapped up 7 months interning with @pcastr.bsky.social at Google DeepMind and I'm so excited to share our work: arxiv.org/abs/2602.10324.
TLDR: We used LLM-powered program synthesis to automatically model and discover differences between human and LLM strategic behavior
TLDR: We used LLM-powered program synthesis to automatically model and discover differences between human and LLM strategic behavior
This is very much why Full Bloom had its bad robots (powered by LLM garbage) and its good robot (powered by a magical consciousness). I will not let the tech bros have a monopoly on robots.
As a science fiction writer I really fucking resent tech bros for ruining the phrase "artificial intelligence." 80 years of it being all about cool androids and interesting computers destroyed by five years of "Spicy autocorrect making everything shittier for money." NEVER FORGIVE.
May 22, 2026 at 11:39 PM
This is very much why Full Bloom had its bad robots (powered by LLM garbage) and its good robot (powered by a magical consciousness). I will not let the tech bros have a monopoly on robots.
ICYMI! Apple is reportedly developing an upgraded Siri, powered by advanced large language models (LLMs), referred to internally as "LLM Siri."
November 22, 2024 at 5:07 PM
ICYMI! Apple is reportedly developing an upgraded Siri, powered by advanced large language models (LLMs), referred to internally as "LLM Siri."
llm-scraper-worker: Turn any webpage into structured data using LLMs
---
A port of llm-scraper to Cloudflare Workers, powered by the browser rendering api and ai sdk.
www.npmjs.com/package/llm-...
---
A port of llm-scraper to Cloudflare Workers, powered by the browser rendering api and ai sdk.
www.npmjs.com/package/llm-...
llm-scraper-worker
llm-scraper-worker. Latest version: 0.0.3, last published: a few seconds ago. Start using llm-scraper-worker in your project by running `npm i llm-scraper-worker`. There are no other projects in the n...
www.npmjs.com
January 7, 2025 at 2:36 PM
llm-scraper-worker: Turn any webpage into structured data using LLMs
---
A port of llm-scraper to Cloudflare Workers, powered by the browser rendering api and ai sdk.
www.npmjs.com/package/llm-...
---
A port of llm-scraper to Cloudflare Workers, powered by the browser rendering api and ai sdk.
www.npmjs.com/package/llm-...
I am ready to question all of the big name indie rpg groups that bought into a LLM powered rpg streaming website with VC funding, and an owner who is an active-duty military strategist tbh
March 18, 2025 at 11:06 PM
I am ready to question all of the big name indie rpg groups that bought into a LLM powered rpg streaming website with VC funding, and an owner who is an active-duty military strategist tbh
Here I was, about to blame ChatGPT, and it was basically a bespoke, steam-powered LLM that did it.
June 14, 2026 at 7:19 PM
Here I was, about to blame ChatGPT, and it was basically a bespoke, steam-powered LLM that did it.
1/ 🚀🧪🤖 Big update for #Omega — the LLM-powered autonomous agent for bioimage analysis in @napari.org ! Custom LLM endpoints, layer actions, plugin extensibility, agentic widget maker, modernized UI, & 30+ fixes. Here's what's new in version *2026.2.8.2*:
February 9, 2026 at 6:40 AM
1/ 🚀🧪🤖 Big update for #Omega — the LLM-powered autonomous agent for bioimage analysis in @napari.org ! Custom LLM endpoints, layer actions, plugin extensibility, agentic widget maker, modernized UI, & 30+ fixes. Here's what's new in version *2026.2.8.2*:
We're like five years away from LLM-powered appliances getting hooked on drugs.
June 2, 2025 at 3:05 AM
We're like five years away from LLM-powered appliances getting hooked on drugs.
Fuck it! Structured Generation w/ SmolLM2 running in browser & WebGPU 🔥
Powered by MLC Web-LLM & XGrammar ⚡
Define a JSON schema, Input free text, get structured data right in your browser - profit!!
Powered by MLC Web-LLM & XGrammar ⚡
Define a JSON schema, Input free text, get structured data right in your browser - profit!!
November 28, 2024 at 10:24 PM
Fuck it! Structured Generation w/ SmolLM2 running in browser & WebGPU 🔥
Powered by MLC Web-LLM & XGrammar ⚡
Define a JSON schema, Input free text, get structured data right in your browser - profit!!
Powered by MLC Web-LLM & XGrammar ⚡
Define a JSON schema, Input free text, get structured data right in your browser - profit!!