fp.

Dual Axes, Not Dueling

https://dbtcharts.com/blog/dual-axes-not-dueling/

1•infowetrust•27s ago•0 comments

Build Self-Improving AI Agents with Temporal and Oodle

https://blog.oodle.ai/self-improving-agent-loops/

1•kirankgollu•1m ago•0 comments

Artificial Analysis Decision Index v1.0

https://artificialanalysis.ai/articles/artificial-analysis-decision-index

1•theanonymousone•1m ago•0 comments

Branded

https://www.imdb.com/title/tt1368440/

1•mahirsaid•2m ago•1 comments

TR-Core – A non-neural symbolic AI architecture built in pure Python

https://github.com/Mardon-programm/tabula-rasa-core

1•Mr_Mardon•2m ago•0 comments

You better start believing in philosophical thought experiments

https://neurobiology.substack.com/p/you-better-start-believing-in-philosophical

1•paulpauper•6m ago•0 comments

Anthropic: The Agony and the Ecstasy

https://www.exponentialview.co/p/anthropic-the-agony-and-the-ecstasy

1•swolpers•7m ago•0 comments

Georgia man sues Flock, sheriff's office over alleged ALPR-fueled retaliation

https://arstechnica.com/tech-policy/2026/10/georgia-man-sues-flock-sheriffs-office-over-alleged-a...

2•nobody9999•8m ago•0 comments

Nunc Sonat, control Apple TV and HomePod music from the Mac menu bar

https://diligiteiustitiam.world/nuncsonat/

1•kobit•9m ago•0 comments

JEV: Type Safe AI

https://typesafe.ai/

2•Betelbuddy•10m ago•3 comments

PhantomWave End to End Encrypted Digital Voice Chat [video]

https://www.youtube.com/watch?v=Ssze-T00Yzk

1•austinallegro•10m ago•0 comments

Pessimists Archive

https://pessimistsarchive.org

1•andsoitis•10m ago•0 comments

Danish Supreme Court Finds That AI CSAM Is CSAM

https://domstol.fe1.tangora.com/Domsoversigt-(H%C3%B8jesteret).31478.aspx?recordid31478=2930

1•delusional•10m ago•2 comments

Fintech startup Ramp hits $60B valuation after $1.85B funding round

https://www.reuters.com/technology/fintech-firm-ramp-raises-185-billion-60-billion-valuation-2026...

1•doppp•10m ago•0 comments

Making a Software Trailer with Astra

https://pierce.dev/notes/making-a-software-trailer-with-astra

2•icyfox•11m ago•0 comments

Show HN: Relay – A Mac app that carries replies between Claude, Codex and Cursor

https://kanary.to/agents

1•kenn•12m ago•0 comments

Opsin Raises $21M to Expand Its Risk Protection Layer

https://www.opsinsecurity.com/blog/opsin-raises-21m-to-expand-its-risk-protection-layer

1•ShayNehmad•12m ago•1 comments

Bill Gates returns to wage war on AI risks

https://www.axios.com/2026/10/15/bill-gates-ai-safety-push

3•samizdis•12m ago•1 comments

AMD Helion for Agentic AI and Enterprise

https://ir.amd.com/news-events/press-releases/detail/1301/introducing-amd-helion-agentic-ai-compu...

1•endangeredhuman•12m ago•0 comments

Strategy to Reduce Subscription Churn

https://thesupportqueue.substack.com/p/strategy-to-reduce-subscription-churn

1•vinothkumarnaga•12m ago•0 comments

OCC Releases Report on Large Banks' Debanking Activities

https://www.occ.gov/news-issuances/news-releases/2026/nr-occ-2026-90.html

1•sparrish•13m ago•0 comments

Package managers have an integrity problem

https://blog.trailofbits.com/2026/10/14/package-managers-have-an-integrity-problem/

2•abhisek•14m ago•0 comments

Agent Vault

https://agentvault.io/

4•reinsbr•14m ago•3 comments

Apple Joins Nvidia in the $5T Valuation Club

https://www.wsj.com/tech/seventh-time-is-the-charm-apple-closes-above-5-trillion-market-capitaliz...

4•bookofjoe•14m ago•1 comments

Preparing for the 2026 US Midterms

https://www.anthropic.com/news/2026-us-midterms

3•swolpers•17m ago•0 comments

Loment 0.1.4.5: Tiered GC and Windows Installer Fix

https://fujojtop.github.io/loment-site/

1•FujoLoment•18m ago•0 comments

Has China lost 130M people?

https://www.thetimes.com/world/asia/article/china-falling-population-130-million-nh6jbcftf

4•randycupertino•19m ago•0 comments

Training a 180B-parameter model on a single 8×H100 node

https://physionlabs.ai/blog/nextron

1•ljhnick•22m ago•0 comments

Researchers use AI to design proteins that cut up other protein

https://arstechnica.com/science/2026/10/researchers-use-ai-to-design-proteins-that-cut-up-other-p...

4•dverlaeckt80•22m ago•0 comments

Show HN: BlastLoop, a structure-first coding method for AI agents

https://github.com/rlshaw-ml/BlastLoop

2•rossleeshaw•23m ago•0 comments

Open in hackernews

Neural Networks Are Already Clean

https://already-clean.goodfire.com/

21•themgt•46m ago

Comments

phyzome•14m ago

16,000 words into the essay: "What I mean by clean"

...and it still doesn't answer it. What is this actually about?

lucrbvi•13m ago

It means that modern neural networks are way more interpretable than older ones. We can understand how they work way more easily than earlier ones.

BenoitP•5m ago

For context: Goodfire has expertise in interpretability through a Sparse Auto Encoder technique of their own

https://www.goodfire.com/blog/a-practical-guide-to-saes