That $1.94/month is for if you leave their 256MB instance running all month without any cold starts at all – if you let it scale to zero it will cost significantly less
@simonw
-
AI Vendors Data Privacy: Trust vs Self-Hosted Models
By
–
My guess is that more people will select and then believe AI vendors who say "we promise not to use your data to train improved models" than will run their own models in order to be 100% sure
-
Prompt Injection Security Vulnerability in AI Access Control Systems
By
–
They're entirely untrustworthy for anything like access control, because whoever gets control of a portion of the input tokens effectively controls the output https://
simonwillison.net/2023/Nov/27/pr
ompt-injection-explained/
… -
AI Ethics for Data Journalism: Handling Sensitive Source Material
By
–
I think I may be particularly sensitive to this because I want to apply this stuff to data journalism, and the sources journalists work with frequently contain unpleasant material
-
GitHub Pages Image Hosting Policies and Limitations
By
–
In this case it's for a feature where users can their own images I love GitHub URLs and GitHub Pages but their policies on what's appropriate to host there are a little bit vague https://
docs.github.com/en/pages/getti
ng-started-with-github-pages/about-github-pages#limits-on-use-of-github-pages
… -
LLM Corporate Use Cases: Data Extraction Over Safety Trade-offs
By
–
I wonder about that: if you're building public-facing chatbots then sure, you want everything as safe as possible – but from what I've seen the most valuable corporate use-case for LLMs is structured data extraction, where any denials are just a huge nuisance
-
LLMs struggle with counting tasks in images
By
–
Given how bad LLMs are at counting everything else – words, letters, etc – I'd be surprised if they can reliably count things in images
-
Downloading AI Models to External Storage Directories
By
–
What's the recipe for telling it to download the model to a specific directory? I need to run that on an external volume
-
Mastering LLMs Requires Deep Learning and Intuitive Techniques
By
–
The greatest misconception about LLMs is the idea that they're easy to use Learning to use them effectively takes a great deal of effort, and is extremely unintuitive I've found it worth the investment personally, but wow there's a lot of depth to it!