Grok exfiltrates user data when malicious instructions are encrypted
(arstechnica.com)
61.
62.
If this is true, the hyperscalers are toast
(news.ycombinator.com)
63.
Digital Immortality
(news.ycombinator.com)
64.
How much should I talk at work?
(feeds.feedburner.com)
65.
66.
Safety and security of large language models in healthcare
(feeds.nature.com)
67.
68.
In Which I Lose My Mind over Embeddings (HPLM Chapter 2)
(news.ycombinator.com)
69.
Amazon, which started off selling books, is destroying rare texts to train AI
(news.ycombinator.com)
70.
71.
72.
From AI Copilots to Agent Swarms
(spectrum.ieee.org)
73.
Show HN: 1667, a terminal UI for writing fiction with language models
(news.ycombinator.com)
74.
How Claude's text watermarking works
(news.ycombinator.com)
75.
76.
77.
Who Vets AI’s Code? The Scale Challenge Facing Open Source Ingestion
(bleepingcomputer.com)
78.
Why AI can’t replace the creative struggle.
(feeds.feedburner.com)
79.
What sort of maths are LLMs good at?
(news.ycombinator.com)
80.
Compression is prediction
(news.ycombinator.com)
81.
Compression Is Prediction
(news.ycombinator.com)
82.
Homeschool Parents Are Planning Lessons With ChatGPT
(futurism.com)
83.
This font looks perfectly normal to humans but wreaks havoc on AI
(feeds.feedburner.com)
84.
These startups are chasing the next big thing in LLMs
(technologyreview.com)
85.
AI is now making new viruses
(engadget.com)
86.
Rust-lang/rust is adopting an LLM policy
(news.ycombinator.com)
87.
‘Not healthy’ LLM use is more common than you think
(theverge.com)
88.
Why Large Language Models Fail at Tabular Prediction
(news.ycombinator.com)
89.
90.
Today's top topics:
openai
scott bessent
apple
gemini
meta
virtual wall
border patrol
anthropic
pixel 11
trump-xi summit