Scott Lindsey
@scottlindsey.bsky.social
Dad / Developer with a boring job, Liberal, Anti-Crypto AI enthusiast
Pinned
Scott Lindsey
@scottlindsey.bsky.social
· Dec 29
Reposted by Scott Lindsey
the one thing that slightly irritates me about this is that it essentially privileges private datasets over public ones
obviously "public dataset of all books ever" is a plain violation of copyright but it would also be way better for the world than book datasets being limited to big companies
obviously "public dataset of all books ever" is a plain violation of copyright but it would also be way better for the world than book datasets being limited to big companies
June 24, 2025 at 8:19 PM
the one thing that slightly irritates me about this is that it essentially privileges private datasets over public ones
obviously "public dataset of all books ever" is a plain violation of copyright but it would also be way better for the world than book datasets being limited to big companies
obviously "public dataset of all books ever" is a plain violation of copyright but it would also be way better for the world than book datasets being limited to big companies
This seems like a really good ruling.
Training is legal, but pirating piles of content remains illegal.
And, verbatim responses expose the LLM provider to liability.
Training is legal, but pirating piles of content remains illegal.
And, verbatim responses expose the LLM provider to liability.
🚨BREAKING: Federal judge concludes that using copyrighted works to train generative A.I. is transformative and ultimately a fair use. (Nevertheless, Anthropic can’t beat the lawsuit because it pirated books for another purpose too.) First of kind ruling. www.documentcloud.org/documents/25...
Bartz
www.documentcloud.org
June 24, 2025 at 9:30 PM
This seems like a really good ruling.
Training is legal, but pirating piles of content remains illegal.
And, verbatim responses expose the LLM provider to liability.
Training is legal, but pirating piles of content remains illegal.
And, verbatim responses expose the LLM provider to liability.
Reposted by Scott Lindsey
Reposted by Scott Lindsey
I think people on this site need to see this.
June 18, 2025 at 3:37 AM
I think people on this site need to see this.
Great piece about AI on NYT Magazine. As usual, Bluesky responses do not disappoint.
June 17, 2025 at 4:06 PM
Great piece about AI on NYT Magazine. As usual, Bluesky responses do not disappoint.
Reposted by Scott Lindsey
The NYT Magazine asked me and @caseynewton.bsky.social to open their AI issue by talking about how we use AI, why it's not going away, and how you can't be a serious critic if you're in denial about how useful it is. Bluesky's gonna love this one!
www.nytimes.com/2025/06/16/m...
www.nytimes.com/2025/06/16/m...
June 17, 2025 at 2:35 PM
The NYT Magazine asked me and @caseynewton.bsky.social to open their AI issue by talking about how we use AI, why it's not going away, and how you can't be a serious critic if you're in denial about how useful it is. Bluesky's gonna love this one!
www.nytimes.com/2025/06/16/m...
www.nytimes.com/2025/06/16/m...
Reposted by Scott Lindsey
It's very funny to watch people with computer science knowledge deny the existence of problems where potential solutions can be trivially validated, e.g. bsky.app/profile/sear...
Closed-loop scenarios come up all the time. I often ask ChatGPT for e.g. "What do you call a part that attaches an X to a Y at a right angle? Give me five options", then check the Lowes website to see which of those is the one I'm actually looking for
May 2, 2025 at 4:37 PM
It's very funny to watch people with computer science knowledge deny the existence of problems where potential solutions can be trivially validated, e.g. bsky.app/profile/sear...
Reposted by Scott Lindsey
enlightenment is realizing that for any individual person, almost all contexts are contexts where they have low expertise
there are 100 different ways your life could be better if you Just Start and claude can make a reasonable starting plan for all of them
so what are you waiting for?
so what are you waiting for?
what you really need is to just start
May 2, 2025 at 4:29 PM
enlightenment is realizing that for any individual person, almost all contexts are contexts where they have low expertise
ChatGPT has largely replaced Google for me.
May 1, 2025 at 11:20 PM
ChatGPT has largely replaced Google for me.
The replies on this are overwhelmed with wish-casting by people afraid of change.
If you can’t find a use for LLMs, that’s you.
They may not be easy to use but for people who can make use of them they are incredibly useful.
If you can’t find a use for LLMs, that’s you.
They may not be easy to use but for people who can make use of them they are incredibly useful.
There are a lot of critiques of LLMs that I agree with but "they suck and aren't useful" doesn't really hold water.
I understand people not using them because of social, economic, and environmental concerns. And I also understand people using them because they can be very useful.
Thoughts?
I understand people not using them because of social, economic, and environmental concerns. And I also understand people using them because they can be very useful.
Thoughts?
April 24, 2025 at 4:07 PM
The replies on this are overwhelmed with wish-casting by people afraid of change.
If you can’t find a use for LLMs, that’s you.
They may not be easy to use but for people who can make use of them they are incredibly useful.
If you can’t find a use for LLMs, that’s you.
They may not be easy to use but for people who can make use of them they are incredibly useful.
Reposted by Scott Lindsey
No one I've talked to or have heard on TV today, could explain how Donald Trump came up with today's tariff numbers.
I checked the other platform and renowned business writer James Surowiecki offered this. It makes sense as an explanation less so for economic policy #Trump #Tariffs
I checked the other platform and renowned business writer James Surowiecki offered this. It makes sense as an explanation less so for economic policy #Trump #Tariffs
April 3, 2025 at 1:21 AM
* make it sunny
* add some people
* make this a travel poster for the Kinsol Trestle
* add some people
* make this a travel poster for the Kinsol Trestle
March 28, 2025 at 2:45 PM
* make it sunny
* add some people
* make this a travel poster for the Kinsol Trestle
* add some people
* make this a travel poster for the Kinsol Trestle
Reposted by Scott Lindsey
Reposted by Scott Lindsey
Here's the table of contents for my lengthy new piece on how I use LLMs to help me write code simonwillison.net/2025/Mar/11/...
March 11, 2025 at 2:11 PM
Here's the table of contents for my lengthy new piece on how I use LLMs to help me write code simonwillison.net/2025/Mar/11/...
Reposted by Scott Lindsey
part of the divide we’ve been seeing in AI discourse is due to the fact that the general public has only experienced messing around with lower-tier free models — they’re not experiencing the flagship models until quite a bit later so their mental model of the intelligence level is lagging
March 4, 2025 at 3:34 AM
part of the divide we’ve been seeing in AI discourse is due to the fact that the general public has only experienced messing around with lower-tier free models — they’re not experiencing the flagship models until quite a bit later so their mental model of the intelligence level is lagging
Reposted by Scott Lindsey
Ok, quick thread on why DeepSeek is blowing up assumptions about LLMs and maybe the whole (U.S.) AI industry. DeepSeek is an AI lab funded by a Chinese hedge fund, their AI code is open source, meaning freely available to use, which most big U.S. models (aside from Meta’s) are not. That’s the start…
January 27, 2025 at 1:04 PM
Ok, quick thread on why DeepSeek is blowing up assumptions about LLMs and maybe the whole (U.S.) AI industry. DeepSeek is an AI lab funded by a Chinese hedge fund, their AI code is open source, meaning freely available to use, which most big U.S. models (aside from Meta’s) are not. That’s the start…
He’s right in principle but wrong in practice. LLMs can’t write books of any quality.
And when they can you will not “stock” them.
The value of an LLM generated book would be the ability to create one on demand to your own highly whimsical requirements.
And when they can you will not “stock” them.
The value of an LLM generated book would be the ability to create one on demand to your own highly whimsical requirements.
I’d stock a book written by AI.
Insistence on the creative as an output only possible from the Special Human Mind is a type of ableism.
Insistence on the creative as an output only possible from the Special Human Mind is a type of ableism.
January 25, 2025 at 4:53 AM
He’s right in principle but wrong in practice. LLMs can’t write books of any quality.
And when they can you will not “stock” them.
The value of an LLM generated book would be the ability to create one on demand to your own highly whimsical requirements.
And when they can you will not “stock” them.
The value of an LLM generated book would be the ability to create one on demand to your own highly whimsical requirements.
Reposted by Scott Lindsey
Happy 25th Anniversary to those who remember...
December 31, 2024 at 8:36 PM
Happy 25th Anniversary to those who remember...
Reposted by Scott Lindsey
I think everyone who has an opnion, good or bad, about LLMs, should read how @simonwillison.net has summer up what’s happened in the space this year. He’s the most credible, most independent, most honest, and most technically fluent person watching the space. simonwillison.net/2024/Dec/31/...
Things we learned about LLMs in 2024
A lot has happened in the world of Large Language Models over the course of 2024. Here’s a review of things we figured out about the field in the past …
simonwillison.net
December 31, 2024 at 9:37 PM
I think everyone who has an opnion, good or bad, about LLMs, should read how @simonwillison.net has summer up what’s happened in the space this year. He’s the most credible, most independent, most honest, and most technically fluent person watching the space. simonwillison.net/2024/Dec/31/...
Reposted by Scott Lindsey
Some folks are pretty good seeing things coming.
December 18, 2024 at 11:37 AM
Some folks are pretty good seeing things coming.
Reposted by Scott Lindsey
One of the most important debates in tech right now is the group of folks who think AI is fake and sucks vs. the people who think AI is real and dangerous. I wrote about why I'm in the latter camp, and talked about my differences with Gary Marcus www.platformer.news/ai-skeptics-...
December 6, 2024 at 1:33 AM
One of the most important debates in tech right now is the group of folks who think AI is fake and sucks vs. the people who think AI is real and dangerous. I wrote about why I'm in the latter camp, and talked about my differences with Gary Marcus www.platformer.news/ai-skeptics-...
How it started / how it’s going, etc.
November 28, 2024 at 9:21 PM
How it started / how it’s going, etc.
Reposted by Scott Lindsey
Some notes on running the new SmolVM vision model (~4.2GB) on my MacBook Pro M2 at 75 tokens/second to describe a photo, using a shell one-liner powered by uv and mlx-vlm simonwillison.net/2024/Nov/28/...
SmolVLM—small yet mighty Vision Language Model
I've been having fun playing with this new vision model from the Hugging Face team behind [SmolLM](https://simonwillison.net/2024/Nov/2/smollm2/). They describe it as: > [...] a 2B VLM, SOTA for its m...
simonwillison.net
November 28, 2024 at 8:30 PM
Some notes on running the new SmolVM vision model (~4.2GB) on my MacBook Pro M2 at 75 tokens/second to describe a photo, using a shell one-liner powered by uv and mlx-vlm simonwillison.net/2024/Nov/28/...
Holy cow the replies on this.
The disconnect is real. You can’t post it and also choose who reads it.
The disconnect is real. You can’t post it and also choose who reads it.
First dataset for the new @huggingface.bsky.social @bsky.app community organisation: one-million-bluesky-posts 🦋
📊 1M public posts from Bluesky's firehose API
🔍 Includes text, metadata, and language predictions
🔬 Perfect to experiment with using ML for Bluesky 🤗
huggingface.co/datasets/blu...
📊 1M public posts from Bluesky's firehose API
🔍 Includes text, metadata, and language predictions
🔬 Perfect to experiment with using ML for Bluesky 🤗
huggingface.co/datasets/blu...
bluesky-community/one-million-bluesky-posts · Datasets at Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
November 27, 2024 at 8:00 PM
Holy cow the replies on this.
The disconnect is real. You can’t post it and also choose who reads it.
The disconnect is real. You can’t post it and also choose who reads it.