doorss

Friday, March 27, 2026 — 16 items

AI & LLMs

Simon Willison / Hacker News · ngrok.com · comments
Sam Rose published another excellent interactive essay, this time explaining how LLM quantization works from first principles. A great resource for understanding how models are compressed for efficient inference.
Lobsters · refactoringenglish.com · comments
An exercise in distinguishing AI-written from human-written design documents, offering insight into the telltale patterns and tropes of LLM-generated prose. Relevant for anyone trying to detect AI writing in professional contexts.
Hacker News · www.symbolica.ai · comments
Symbolica achieved 36% on the new ARC-AGI-3 benchmark on its first day, using program synthesis approaches. This is a significant result on a benchmark designed to test general reasoning beyond pattern matching.
Hacker News · www.reco.ai · comments
Reco used AI to port JSONata from JavaScript to Go in a single day, eliminating a costly dependency and saving $500K annually. A compelling case study of 'vibe porting' — using LLMs to translate entire codebases between languages.
Hacker News · www.claudescode.dev · comments
Analysis shows the vast majority of Claude Code-linked commits go to tiny, low-star repos, suggesting AI coding tools are predominantly used for personal and small-scale projects rather than major open source work.

Privacy & Security

Lobsters · arxiv.org · comments
Research paper demonstrating how LLMs can be used to deanonymize online users at scale, posing serious implications for privacy. This intersects both AI capabilities and the erosion of online anonymity.
Hacker News · fightchatcontrol.eu · comments
The EU's Chat Control proposal continues to push for client-side scanning of private messages and photos, raising major concerns about mass surveillance and encryption undermining. The campaign urges citizens to push back.
Hacker News · calpaterson.com · comments
A clear explanation of prompt injection attacks where adversarial inputs instruct LLMs to ignore prior instructions. Highlights a fundamental and still-unsolved security problem in deploying LLMs in production systems.

Advertising & Platform Business

Financial Times · www.ft.com
A US judge dismissed Musk's lawsuit alleging that major brands organized an illegal boycott of X (formerly Twitter). The ruling is a blow to Musk's attempts to force advertisers back onto the platform.
Hacker News · colibri.social · comments
Colibri is a new community chat platform built on Bluesky's AT Protocol, extending the decentralized social web beyond microblogging into group messaging and community spaces.

Anthropic & AI Industry

Financial Times · www.ft.com
A judge blocked the Trump administration from classifying Anthropic as a 'supply chain risk,' calling the designation 'arbitrary and capricious.' The case stems from Anthropic's dispute with the Pentagon over AI deployment.
Financial Times · www.ft.com
OpenAI is showing more business discipline by ditching the Sora video app and other consumer experiments, refocusing on its core enterprise and developer products.

Software Supply Chain & Security Incidents

Hacker News · futuresearch.ai · comments
Callum McMahon shares a detailed timeline of how he discovered and reported the LiteLLM supply chain attack to PyPI. A gripping incident response narrative highlighting the fragility of the Python package ecosystem.
Lobsters · daniel.haxx.se · comments
Daniel Stenberg (curl maintainer) argues that blind trust in software dependencies is dangerous and lays out practical approaches for verifying the software you depend on.

Suomi

Yle pääuutiset · yle.fi
Kela lopetti kokeilun, jossa tekoälyllä yritettiin tunnistaa väärinkäytöksiä. Syynä oli epäselvyys siitä, mitä algoritmilla saa laillisesti tehdä – Ruotsissa vastaava järjestelmä syrji tiettyjä ryhmiä.
Yle pääuutiset · yle.fi
IT-alan korkeakouluopintoihin hakeneiden määrä laski yli 18 prosenttia tämän kevään yhteishaussa. Taideopinnot sen sijaan kasvattivat suosiotaan.
My preferences