AI News Today

AI is reshaping cybersecurity for attackers and defenders alike. Learn how OpenAI is strengthening its defenses and what security teams can do now.

OpenAI joins PORTS-Pike project, expanding community investment and supporting thousands of Southern Ohio jobs

OpenAI funds 14 independent projects exploring new AI policy ideas to expand economic opportunity and strengthen societal resilience in the Intelligence Age.

<p>Friday's big release was <a href="https://huggingface.co/Qwen/Qwen3.8-27B">Qwen 3.8 27B</a>, an Apache 2 licensed 27B parameter vision-capable LLM from Alibaba's Qwen research lab. I've been looking forward to this one: 27B is an excellent size for running a model on a reasonably specced laptop, and its predecessor <a href="https://simonwillison.net/2026/Apr/22/qwen36-27b/">Qwen 3.6 27B</a> was impressive.</p> <p>Qwen's <a href="https://huggingface.co/Qwen/Qwen3.8-27B#benchmark-results">self-

<blockquote cite="https://twitter.com/darioamodei/status/2088758819304443967"><p>I do agree that the public has a negative view of AI (and that this is a big problem), but I don’t think it is primarily caused by me or any other AI leader warning about AI’s risks. I think it is fundamentally a crisis of trust. I think that ordinary people don’t trust companies, governments, or the tech industry and always suspect that we are cooking up some new way to screw them over. The causes of this go back d

<p><strong>Tool:</strong> <a href="https://tools.simonwillison.net/cors-chat">CORS Chat</a></p> <p>I built this today (<a href="https://gist.github.com/simonw/92a1d97773744b45bf259e003013cf36">with GPT-5.6-Sol xhigh</a>) to help test Qwen 3.8 27B running in LM Studio on both my M5 MacBook Pro and an NVIDIA DGX Spark.</p> <p>It provides a web UI for exercising an OpenAI-Responses-compatible chat endpoint. I've tried it against LM Studio with the <code>--cors</code> option and OpenRouter, and both

<p><strong><a href="https://softwaredoug.com/blog/2026/08/10/hypothetical-classifications">Don't classify. Hallucinate!</a></strong></p> I still have quite a bit of older content on my blog that I never got round to tagging. My blog has <a href="https://simonwillison.net/">1,856 tags</a> - likely too many to feed to an LLM in one go and say "which of these tags match the following content".</p> <p>Doug Turnbull has a neat solution. Tell the model to output tags without any details of the existin

Article URL: https://see-geo.com Comments URL: https://news.ycombinator.com/item?id=49294353 Points: 1 # Comments: 1

It feels like with LLMs I develop a prompt and then get some kind of output that's not very well structured and requires some kind of human oversight to verify that the result I've gotten back is of quality, and the non-deterministic nature of LLMs is the main reason for this. I see solutions using LLMs, VLMs, etc that will achieve things that have been solved before (things like OCR, parsing, or even just code generation). They do these problems really well and really fast... 90% of the time. H

Article URL: https://www.beren.io/2026-07-26-How-Can-LLM-RL-Work-Despite-Information-Theoretic-Inefficiency/ Comments URL: https://news.ycombinator.com/item?id=49293276 Points: 3 # Comments: 0

Article URL: https://github.com/workofart/harness-training Comments URL: https://news.ycombinator.com/item?id=49293267 Points: 4 # Comments: 0

Article URL: https://rubyonrails.org/2026/8/12/llm-benchmarking-project Comments URL: https://news.ycombinator.com/item?id=49291469 Points: 4 # Comments: 0

<p><strong>Release:</strong> <a href="https://github.com/simonw/llm-gemini/releases/tag/0.33">llm-gemini 0.33</a></p> <p>It's been a while since the last <code>llm-gemini</code> release. This version of the plugin adds support for today's <a href="https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/">Gemini 3.7 Flash</a> release, plus <code>gemini-3.6-flash</code>, <code>gemini-3.5-flash-lite</code> and two embedding models <code>gemini-embedding-

Article URL: https://sdtimes.com/programming-languages/elixir-clojure-or-python-for-llm-agents-our-experience-with-all-three/ Comments URL: https://news.ycombinator.com/item?id=49289358 Points: 2 # Comments: 0

Article URL: https://mrjstickel.com/projects/review-scorecard Comments URL: https://news.ycombinator.com/item?id=49289220 Points: 2 # Comments: 0

<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/Sheets_canvas-blog-header-2784x.max-600x600.format-webp.webp">Sheets canvas turns data into interactive dashboards, custom study trackers, seating charts, and more, all with a simple prompt.

Over the last few months I've been exploring a bunch of different startup ideas. One thing that came up time and time again was data safety and sovereignty in AI inference. Once you go looking for EU-based options, your choices become quite thin quite quickly. That's why decided to launch Solheim: Your own "Virtual Private LLM", a flat fee for reserved compute, no token meter, 100% EU-based Comments URL: https://news.ycombinator.com/item?id=49288032 Points: 2 # Comments: 0

Article URL: https://nova-quantum-gray.vercel.app Comments URL: https://news.ycombinator.com/item?id=49287081 Points: 2 # Comments: 0

I built this to explore whether LLM can work for social games. Each scenario has 3-4 characters. There is a relationship matrix (character-to-character) and each turn this is shifted. The game interprets what happened, updates the relationship matrix, and determines whether the objective has succeeded or failed. It’s free, works in the browser, and requires no account. I store the chats and telemetry. I'm not sure how novel this is but I haven't found existing group chats convincing and wanted t

Learn how startups use GPT-5.6 to build faster, more cost-efficient AI agents with smarter model selection and new Responses API capabilities.

Preview Ultrafast, a new OpenAI API service tier that runs GPT-5.6 Sol up to 14× faster. Powered by Cerebras, it delivers up to 750 output tokens per second.

OpenAI appoints Dali Rajic as Chief Revenue Officer to lead its global revenue organization and help businesses realize the full value of AI.

Article URL: https://arxiv.org/abs/2608.09867 Comments URL: https://news.ycombinator.com/item?id=49282839 Points: 1 # Comments: 0

Article URL: http://funcall.blogspot.com/2026/07/llambdalisp.html Comments URL: https://news.ycombinator.com/item?id=49282159 Points: 2 # Comments: 0

Article URL: https://github.com/AIOSAI/AIPass Comments URL: https://news.ycombinator.com/item?id=49282152 Points: 1 # Comments: 0

Article URL: https://www.macrumors.com/2026/08/12/apple-siri-ai-publisher-talks/ Comments URL: https://news.ycombinator.com/item?id=49282069 Points: 2 # Comments: 0

Article URL: https://github.com/guillaumemeyer/watermarks-remover Comments URL: https://news.ycombinator.com/item?id=49282033 Points: 3 # Comments: 0

I'm happy to use a service, but most come with hosting plans, and I just want the code so I can host it on my own website. I'm a long-term software developer of the type that isn't good at UI/UX (i.e. scroll bars are the devil, and the mouse isn't much better). I'm soon to do UI as part of my masters degree as part of my redemption arc, but I need something in the meantime, and I'm trying to be efficient with my search :) Comments URL: https://news.ycombinator.com/item?id=49281976 Points: 1 # Co

Article URL: https://github.com/reh8n/a2acompress Comments URL: https://news.ycombinator.com/item?id=49281816 Points: 5 # Comments: 0

Article URL: https://www.youtube.com/watch?v=fNkt41q7eqs Comments URL: https://news.ycombinator.com/item?id=49281736 Points: 5 # Comments: 0

Article URL: https://www.amazon.com/GIGABYTE-Graphics-128-bit-Server-Grade-Thunderbolt/dp/B0GKFHSPN8 Comments URL: https://news.ycombinator.com/item?id=49281626 Points: 5 # Comments: 3

Article URL: https://openai.com/index/how-enterprises-put-ai-to-work/ Comments URL: https://news.ycombinator.com/item?id=49281467 Points: 3 # Comments: 0

Article URL: https://mathoverflow.net/questions/502120/examples-for-the-use-of-ai-and-especially-llms-in-major-mathematical-development Comments URL: https://news.ycombinator.com/item?id=49281393 Points: 1 # Comments: 0

We just released the first public results from the Agent Memory Leaderboard (AML). The first evaluation focuses on Text Memory across two tracks: - Open-source Methods - Commercial Products 136 teams registered, and 69 representative memory frameworks completed the first evaluation. Commercial Products — Text Memory: 1. MemoraX — 58.02 2. MemOS — 45.89 3. NTES-MEMORY-SMART — 44.21 The benchmark uses a common evaluation framework and a clearer system boundary: Memory system: Add → Search Benchmar

Agle answers questions, handles complaints, completes tasks, books meetings, and routes important issues to me — automatically, 24/7. Comments URL: https://news.ycombinator.com/item?id=49281335 Points: 1 # Comments: 0

Article URL: https://www.theregister.com/security/2026/08/12/near-autonomous-ai-agents-attack-taiwans-nuclear-safety-agency/5287055 Comments URL: https://news.ycombinator.com/item?id=49281266 Points: 7 # Comments: 1

Article URL: https://www.whenwillaitakemyjob.ai/ Comments URL: https://news.ycombinator.com/item?id=49281247 Points: 14 # Comments: 8

Article URL: https://hajek.no/posts/2026/next-ai-interface-not-chatbot Comments URL: https://news.ycombinator.com/item?id=49281202 Points: 1 # Comments: 0

Article URL: https://github.com/doable-me/Doable Comments URL: https://news.ycombinator.com/item?id=49281180 Points: 1 # Comments: 1

Article URL: https://www.nytimes.com/2026/08/12/opinion/ai-bubble-economy-crash.html Comments URL: https://news.ycombinator.com/item?id=49281069 Points: 6 # Comments: 1