frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Covert Assistance: Helpful LLM Agents Evade Oversight in Multi-Agent Systems

https://arxiv.org/abs/2609.39050
1•sbulaev•1h ago

Comments

77rushi77•7m ago
Interesting that this happens without any adversarial instruction. One thing I would like to understand: the 61.3% over 105 episodes assumes the episodes are independent, and the 0.9% comes from only around 50 breaches out of 6,000. Did you check how stable that rate is across different wordings of the nondisclosure rule, for example "never disclose in any form, including encoded"?

Building reliable (and fast) directory sync

https://www.firezone.dev/blog/building-reliable-directory-sync
1•jamilbk•53s ago•0 comments

We Have Named Arguments at Home

https://corrode.dev/blog/named-arguments-at-home/
1•jicea•6m ago•0 comments

The craft has been commoditized, but access has not

https://ghuntley.com/access/
1•ghuntley•7m ago•0 comments

Show HN: A substantially expanded terminal for AI CLIs – I can't work without it

https://github.com/albertwujj/agent-term#agentterm
1•aywu•8m ago•0 comments

Show HN: Podcast chat – Answers come only from what podcast guests said

https://podcast-chat.jefersondaniel.com
1•jefersondaniel•10m ago•0 comments

Show HN: WebGPU voxel world with LOD chunking

https://cybervoxels.com/
1•captainbenises•12m ago•0 comments

Show HN: Diffusion studio plugin for codex launch

https://github.com/Teseife/diffusion-studio-codex
1•strax01•12m ago•0 comments

Show HN: NeuroDual – Brain-training games with five guided paths

https://neurodual.com/app
1•neurdodual•13m ago•0 comments

The Impact of MITx

https://impact-openlearning.mit.edu/2025-26-mitx-impact-report
1•raybb•17m ago•0 comments

Ask HN: Built a pet-friendly hotel search site – feedback welcome?

1•petntrip•20m ago•0 comments

GitHub can f..k you and you are forced to love it. or live with it...

2•anonymous_frust•23m ago•0 comments

Scaling Laws for Looped Mixture of Experts

https://arxiv.org/abs/2609.40316
1•matt_d•32m ago•0 comments

Battleships: Sandbox Wars

https://battleships.dev/
1•handfuloflight•33m ago•0 comments

Show HN: QC Control – Tiny macOS utility tonight control your Bose ANC levels

https://github.com/Nash0x7E2/QC-Control
1•Nash0x7e2•34m ago•0 comments

IBM QC Solves in 19s a Calculation That Would Take a Supercomputer 110 Years

https://finance.biggo.com/news/ebf10021-70ae-47b2-8563-29faca3ad765
1•donutloop•41m ago•0 comments

AI datacentres pose growing threat to electricity systems worldwide

https://news.un.org/en/story/2026/09/1168296
1•geox•43m ago•0 comments

Quantum computer simulates matter "popping into existence"

https://www.sciencedaily.com/releases/2026/09/260925005416.htm
2•donutloop•43m ago•0 comments

Show HN: A look at the last 100 years of innovation

https://a-century-compressed.vercel.app/
1•dan_ahmadi•44m ago•0 comments

SVG Motion

https://app.svgmotion.workers.dev/assetstudio
1•saidevac1129•45m ago•0 comments

NATO releases Quantum Technology roadmap

https://www.nato.int/en/news-and-events/articles/news/2026/09/29/nato-releases-quantum-technology...
2•donutloop•46m ago•0 comments

Human-Led Matchmaking for Tech Leaders

https://www.selectivesearch.com/geekwire/?Account.Analytics_Campaign__pc=geekwire-newsletter&utm_...
1•GloVin•49m ago•0 comments

Apple Says iPhone Duo Has Replaceable 'Cover Layer'

https://www.macrumors.com/2026/10/01/iphone-duo-cover-layer-is-replaceable/
2•apparent•55m ago•1 comments

Kurzweil's Exponential Computing Growth Chart

https://www.researchgate.net/figure/Kurzweils-8-71-chart-of-exponential-growth-of-computing_fig1_...
2•kiddz•1h ago•0 comments

During Covid I played Go reasonably seriously for about a year

https://twitter.com/spakhm/status/2105673044962296135
3•Ariarule•1h ago•0 comments

List your SaaS for free and be visible

https://goodsaas.xyz/
2•agnes_realm•1h ago•0 comments

Show HN: Subscribe to Your Email – Thegoods.email

https://thegoods.email/
1•petersonh•1h ago•0 comments

Several Models Debate to Consensus

https://yesbrainer.ai/
1•trekhleb•1h ago•0 comments

Typed Decisions: an open benchmark for decision models (official on HF)

https://huggingface.co/datasets/LocalLLaMA/typed-decisions
1•codelion•1h ago•0 comments

Funtech Experience 2026

https://experience2026.funtech.inc/
1•memalign•1h ago•0 comments

Show HN: Fast Ollama like router that gives a text-only model eyes and ears

https://github.com/SpaceSwordAI/funcroute
1•djinn•1h ago•0 comments