Eieye
Eieye
FeedChatReportsAbout
ModelsAgentsCoding toolsOpen sourceResearchInfrastructureAI applicationsSafety and policyBusiness and industry
Filter stories by topic
ModelsAgentsCoding toolsOpen sourceResearchInfrastructureAI applicationsSafety and policyBusiness and industry

Updated 1 Sept, 17:26

6 Aug 2026
ModelsResearch

LLMs won't break symmetric crypto

Anthropic's Claude Mythos found meaningful attacks on the HAWK post-quantum signature candidate and improved a reduced-round AES attack, but poses no threat to deployed AES. The discussion examines whether LLMs can discover new cryptanalytic techniques and where they may instead help find implementation bugs or proof errors.

HN Discussion
5 Aug 2026
ModelsAI applications

Show HN: My receipt printer prints an original artwork every morning

A Raspberry Pi and thermal printer produce a unique daily receipt artwork designed by Claude from weather, dates, and news. The project combines structured model output, constrained retro graphics, and a physical alternative to morning feeds.

HN Discussion
5 Aug 2026
ModelsAgentsCoding toolsBusiness and industry

Muse Code and Muse Spark 1.2

Meta launched Muse Spark 1.2, a coding-focused model, alongside Muse Code, a terminal agent with persistent subagents and crash-recoverable execution. HN discusses its benchmark performance, harness coupling, unusually low data-sharing pricing, and substantial privacy concerns.

HN Discussion
5 Aug 2026

Built by Will Etheridge

wjeth.comwjeth@pm.me
ModelsAgentsResearchInfrastructure

Beating GPT-5.6 Sol on retrieval with 100x cheaper open models

Castform describes RL post-training small open-weight models to outperform frontier models on agentic retrieval over company data at far lower token cost. HN discusses benchmark credibility, retrieval limitations, model specialization, routing, privacy, and the ongoing maintenance cost of adapting to data drift.

HN Discussion
5 Aug 2026
ModelsInfrastructureBusiness and industry

Anthropic Is Building Its Own Chip

Anthropic is building an in-house custom-silicon team to optimize hardware for Claude and reduce reliance on Nvidia and other suppliers. HN discusses whether ASIC efficiency can outweigh slower iteration as AI models and workloads evolve.

HN Discussion
5 Aug 2026
ModelsResearchAI applicationsSafety and policy

I’m leaving OpenAI to build telepathy

A former OpenAI researcher describes Conduit’s plan to decode non-invasive neural signals into text and eventually let AI communicate back into the brain. HN debates the technical plausibility, ambitious timelines, and profound privacy and safety risks of AI-linked thought reading.

HN Discussion
5 Aug 2026
ModelsResearchBusiness and industry

Jeff Dean Leaving Google

Jeff Dean and longtime collaborators are leaving Google to found Discovery Loop, a public-benefit company focused on automating machine learning, science, and engineering to accelerate discovery.

HN Discussion
5 Aug 2026
ModelsResearchBusiness and industry

Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

Google is moving Demis Hassabis from day-to-day DeepMind leadership to Chair and Alphabet Chief Scientist, while Koray Kavukcuoglu takes operational control and Jeff Dean and Sanjay Ghemawat launch Discovery Loop. HN debates whether this marks a strategic shift from blue-sky AGI research toward Gemini commercialization, amid concerns about talent loss and competitiveness.

HN Discussion
← NewerPage 17