News
The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.
| Source | Age | Title | Summary | Status |
|---|---|---|---|---|
hn hn-ai | 1mo | Jujutsu Has No "Create a Branch" Step, and That Made Git Branches Click | — | ↗Not briefed |
hn hn-ai | 1mo | Super Smash Brothers Port to Original Xbox from N64 Decomp | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Epic Games is reportedly pushing its staff to work in an AI-first production model and delaying a secret Marvel and Star Wars crossover sho… | Epic Games is reportedly pushing its staff to work in an AI-first production model and delaying a secret Marvel and Star Wars crossove… | ↗Not briefed |
hn hn-ai | 1mo | LLMs Broke Developer Education | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Interesting data. According to this (caveat emptor), it still is basically OpenAI and Anthropic, everyone else does not matter at all -> | Interesting data. According to this (caveat emptor), it still is basically OpenAI and Anthropic, everyone else does not matter at all… | ↗Not briefed |
hn hn-ai | 1mo | A Real DuckDB Database in S3, with Nothing Else to Run | — | ↗Not briefed |
hn hn-ai | 1mo | Minecraft in Witcher 3 | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | If Anthropic goes public the shareholders should sue for overpaying their Data scientists. | If Anthropic goes public the shareholders should sue for overpaying their Data scientists. | ↗Not briefed |
bluesky bluesky-ai | 1mo | i still have a twitter for like, 10 people i know, no for you page. one of them has a job at anthropic and it’s pretty interesting how ther… | i still have a twitter for like, 10 people i know, no for you page. one of them has a job at anthropic and it’s pretty interesting how… | ↗Not briefed |
bluesky bluesky-ai | 1mo | An instruction to never write any hand-written code again is an instruction to cease developing your software career. You are now an AI ma… | An instruction to never write any hand-written code again is an instruction to cease developing your software career. You are now an… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Monitor Postgres Health | Hello, my name is Danny and I created dbhealth to monitor postgres. I have an SRE background and was working in data eng for a couple… | ↗Not briefed |
hn hn-ai | 1mo | Ssrf-agent 1.0.5 reaches loopback through 2 address spellings | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | 生成AIに「主体」が存在するか否かという話題を眺めていると人間の主体という奈落のような不条理のことを何だと思っているんだ…となるんだけどもしかすると自己の主体の複雑さに対して無自覚だからLLMごときに主体が存在すると錯覚するのだろうか？それはもうむしろ人間の意識がLLM化してい… | 生成AIに「主体」が存在するか否かという話題を眺めていると人間の主体という奈落のような不条理のことを何だと思っているんだ…となるんだけどもしかすると自己の主体の複雑さに対して無自覚だからLLMごときに主体が存在すると錯覚するのだろうか？それはもうむしろ人間の意識がLL… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I don't think mechsplo-themed harnesses will improve LLM performance. Except maybe Gemini, which might be into that shit. | I don't think mechsplo-themed harnesses will improve LLM performance. Except maybe Gemini, which might be into that shit. | ↗Not briefed |
bluesky bluesky-ai | 1mo | I want to study the brain worms Anthropic employees get when they are hired. It's truly fascinating. | I want to study the brain worms Anthropic employees get when they are hired. It's truly fascinating. | ↗Not briefed |
bluesky bluesky-ai | 1mo | at some point when you have a question that doesn’t need the most frontier model you’d be able to just ssh into your router or smart thermo… | at some point when you have a question that doesn’t need the most frontier model you’d be able to just ssh into your router or smart t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Over the last few months, we’ve made significant progress in reducing the effectiveness of distillation attacks against Claude. In this po… | Over the last few months, we’ve made significant progress in reducing the effectiveness of distillation attacks against Claude. In th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | crypto/nft/metaverse was never gonna hit that curve, but llm/ai has/will and many of us will be poorer for it | crypto/nft/metaverse was never gonna hit that curve, but llm/ai has/will and many of us will be poorer for it | ↗Not briefed |
bluesky bluesky-ai | 1mo | honestly I was always of the understanding that thanks to capitalism, all llm/ai has to be is "good enough" to devour huge parts of entire … | honestly I was always of the understanding that thanks to capitalism, all llm/ai has to be is "good enough" to devour huge parts of en… | ↗Not briefed |
bluesky bluesky-ai | 1mo | This is basically the argument that Anthropic has been making with discussions of loop and graphs. They argue you just tell Claude "build c… | This is basically the argument that Anthropic has been making with discussions of loop and graphs. They argue you just tell Claude "bu… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i don’t care who mark sends i’m NOT using an LLM served from meta’s servers | i don’t care who mark sends i’m NOT using an LLM served from meta’s servers | ↗Not briefed |
bluesky bluesky-ai | 1mo | saying avoiding gacha and ai or following bds is too much are always ridiculous because like. even if every single game uses claude or is p… | saying avoiding gacha and ai or following bds is too much are always ridiculous because like. even if every single game uses claude or… | ↗Not briefed |
bluesky bluesky-ai | 1mo | hypno community sucks sometimes. i don't want to be hypnotized by some cold unfeeling ai (large language model), i want to be hypnotized by… | hypno community sucks sometimes. i don't want to be hypnotized by some cold unfeeling ai (large language model), i want to be hypnotiz… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I’ve been a LibreOffice user for years, this only makes me like it more (and other LLM news) The day job had new Cyber Security Training, … | I’ve been a LibreOffice user for years, this only makes me like it more (and other LLM news) The day job had new Cyber Security Train… | ↗Not briefed |
bluesky bluesky-ai | 1mo | just adding to this convo: i perceive a clear difference between studios like Epic openly embracing AI and some indie project using Claude … | just adding to this convo: i perceive a clear difference between studios like Epic openly embracing AI and some indie project using Cl… | ↗Not briefed |
hn hn-ai | 1mo | Setupdoctor: Brew doctor for any repo you just cloned | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic, trying their hardest to both make sure that we go to the Bad Future AND at the same time trying to show their products are simpl… | Anthropic, trying their hardest to both make sure that we go to the Bad Future AND at the same time trying to show their products are… | ↗Not briefed |
hn hn-ai | 1mo | AI-critic – Task-aware auditing of ML models | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | It could, though. A stochastic parrot, if prompted, could absolutely put the english words in the order which a human being would otherwise… | It could, though. A stochastic parrot, if prompted, could absolutely put the english words in the order which a human being would othe… | ↗Not briefed |
bluesky bluesky-ai | 1mo | ...they have to tell investors they are using AI, and how that will improve the product and make them money, but they also have to hide the… | ...they have to tell investors they are using AI, and how that will improve the product and make them money, but they also have to hid… | ↗Not briefed |
rss gary-marcus | 1mo | Could this be the Frontier Labs’ WeWork moment? | Some ominous signs | ↗Not briefed |
hn hn-ai | 1mo | Show HN: RoboGym – open-source (Apache-2.0) Dual arm SO101 playground | After seeing so many demos on web on how Astra works with sim2real, I decided to build my own with dual arm SO101. My physical setup i… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i guess my main feeling on this sort of thing is that it's very clear by now that employees of ai companies routinely lie about the capabil… | i guess my main feeling on this sort of thing is that it's very clear by now that employees of ai companies routinely lie about the ca… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic pays like $400k for data scientists just to get ugly dynamite plots. www.anthropic.com/research/sel... | Anthropic pays like $400k for data scientists just to get ugly dynamite plots. www.anthropic.com/research/sel... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Seems bad. I doubt this is intentional -- this smells like more LLM enshitification -- but it's 100% misleading as to what a government ele… | Seems bad. I doubt this is intentional -- this smells like more LLM enshitification -- but it's 100% misleading as to what a governmen… | ↗Not briefed |
bluesky bluesky-ai | 1mo | if it's helpful to hear, the current cohort of MSU gamedev students _hate_ LLM-based AI tools for just about everything relative to opinion… | if it's helpful to hear, the current cohort of MSU gamedev students _hate_ LLM-based AI tools for just about everything relative to op… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Everyone worries about ChatGPT and Claude when it comes to bio-risk. The AI tools actually powering virus engineering are specialized model…🔧 ChatGPT | Everyone worries about ChatGPT and Claude when it comes to bio-risk. The AI tools actually powering virus engineering are specialized… | ↗Not briefed |
rss techmeme | 1mo | Leading Chinese mathematicians launch an alliance, including eight orgs and aiming for 1,000+ scientists, to focus on developing specialized AI models for math (Chao Kong/South China Morning Post) | Chao Kong / South China Morning Post: Leading Chinese mathematicians launch an alliance, including eight orgs and aiming for 1,000+ sc… | ↗Not briefed |
hn hn-ai | 1mo | Turn a frontier AI into a multi-agent systems architect | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | I’ve been going on about how doomers need more plausible scenarios. I’m glad Anthropic is doing it so we can surface their unstated assumpt… | I’ve been going on about how doomers need more plausible scenarios. I’m glad Anthropic is doing it so we can surface their unstated as… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Wang is Meta’s chief AI officer. Lots of people are reporting bugs and the model not performing as touted by Meta upon release earlier tod… | Wang is Meta’s chief AI officer. Lots of people are reporting bugs and the model not performing as touted by Meta upon release earlie… | ↗Not briefed |
hn hn-ai | 1mo | Chain Inspiring License, a copyleft for the REA era | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: SolidCraft – Open-Source Parametric CAD in Rust (With MCP Support) | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Recover Claude Code Desktop sessions after switching accounts | — | ↗Not briefed |
hn hn-ai | 1mo | Instant Software Factories | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | There is a real human urge to put things into a box labeled with a good or bad category so we can file it away and stop thinking about it. … | There is a real human urge to put things into a box labeled with a good or bad category so we can file it away and stop thinking about… | ↗Not briefed |
bluesky bluesky-ai | 1mo | So, so, so much Hanlon’s razor going on here. | So, so, so much Hanlon’s razor going on here. | ↗Not briefed |
bluesky bluesky-ai | 1mo | This is the most comprehensive description yet of how a model can “escape”. Out requires a few assumptions, like a worm that is exploiting … | This is the most comprehensive description yet of how a model can “escape”. Out requires a few assumptions, like a worm that is exploi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | ColGrep-Agent-2B a local open agent that you can pop into Claude Code or whatever to get cheap local search ColGrep: multivector code sea… | ColGrep-Agent-2B a local open agent that you can pop into Claude Code or whatever to get cheap local search ColGrep: multivector cod… | ↗Not briefed |
hn hn-ai | 1mo | Open Agentic Swarm Security Engineering | — | ↗Not briefed |