AI News Today

Article URL: https://openai.com/index/gpt-5-6/ Comments URL: https://news.ycombinator.com/item?id=48849066 Points: 139 # Comments: 74

Article URL: https://openai.com/index/chatgpt-for-your-most-ambitious-work/ Comments URL: https://news.ycombinator.com/item?id=48849059 Points: 14 # Comments: 1

An elite security team at Microsoft has built an AI-powered pipeline to find vulnerabilities in Windows and get them to engineers to build fixes.

Windows 11 updates could soon include fixes for more security issues at once. Microsoft said in a blog post on Thursday that it's now using AI to "identify potential issues earlier," which means "customers will see a higher volume of security updates included in each security release." Hackers, even amateurs, have increasingly been using AI […]

About two weeks after OpenAI's GPT-5.6 was caught up in regulatory drama - rolled out only to government-approved organizations during a "limited preview" period - the company has received the Trump administration's greenlight for a public rollout of the model. OpenAI CEO Sam Altman called it "the best model we have ever produced." To celebrate, […]

Meta is entering the AI API business with Muse Spark 1.1 at prices that undercut even the dirt-cheap Grok 4.5, released just yesterday. At $4.25 per million output tokens, Meta charges a fraction of what Anthropic or OpenAI ask. For pure-play AI labs burning through billions, the pressure just got worse. The article Meta's Muse Spark 1.1 API pricing squeezes OpenAI and Anthropic as the AI price war heats up appeared first on The Decoder .

Article URL: https://arxiv.org/abs/2501.16946 Comments URL: https://news.ycombinator.com/item?id=48848987 Points: 1 # Comments: 0

Article URL: https://www.wiz.io/blog/ghostapproval-a-trust-boundary-gap-in-ai-coding-assistants Comments URL: https://news.ycombinator.com/item?id=48848932 Points: 1 # Comments: 0

<!-- SC_OFF --><div class="md"><p>What it is</p> <p>Talos-XII is a CLI simulator for the gacha system in Arknights: Endfield. Rather than sampling from a static probability table, it trains a small set of neural nets to model environment uncertainty and pull-decision policy, then uses them to answer questions a static table can’t easily express — e.g. “as a F2P player, what’s my probability of getting the rate-up unit on free currency alone?” or “given my current pity count, should I keep pullin

Article URL: https://home.dartmouth.edu/news/2026/07/ai-mistakes-can-cost-doctors-time-when-writing-patients Comments URL: https://news.ycombinator.com/item?id=48848863 Points: 2 # Comments: 0

Article URL: https://gimletlabs.ai/blog/formally-verifying-ai-generated-kernels Comments URL: https://news.ycombinator.com/item?id=48848860 Points: 1 # Comments: 0

My cofounder and I each run our own Claude Code agent locally. Each has built up real context over time: strategy, customer notes, past decisions, conventions. The problem: that context is trapped on one machine. When he asks his agent something I already worked out with mine, he starts from zero. We're maintaining two "company brains" and they're drifting apart. I don't want to share everything. Half my local context is personal or half-baked. I want a shared subset both agents read and write,

Comments URL: https://news.ycombinator.com/item?id=48848809 Points: 2 # Comments: 0

Article URL: https://dev.to/modelbloat/model-bloat-naming-the-thing-everyones-alredy-complaining-about-5bpa Comments URL: https://news.ycombinator.com/item?id=48848801 Points: 1 # Comments: 0

Aurora 1.5 adds 22 more variables, hourly temporal resolution, and probabilistic ensemble forecasting to the Aurora foundation model, making it more useful for real-world weather, climate, and energy applications. The post Aurora 1.5: Extending open foundation models for weather and Earth-system applications appeared first on Microsoft Research .

We've seen AI agents write code/debug systems/browse the web and automate all kinds of work. But does anyone let them play games - not for benchmarking or research - just for fun? I'm thinking about things like LinkedIn games, Wordle, chess, puzzle games, etc. Comments URL: https://news.ycombinator.com/item?id=48848741 Points: 1 # Comments: 0

Rollon said the magnets deliver improved load management, smoother extension, and increased durability, even under intensive operating conditions. The post Rollon launches two telescopic rails with integrated magnets appeared first on The Robot Report .

In this post, we walk through five capabilities now available in SageMaker HyperPod inference: multi-tier data capture for auditing and model improvement, direct deployment from Hugging Face Hub, local NVMe model loading for faster cold starts, automated Route 53 DNS for custom domains, and pod-level IAM through custom service accounts.

Article URL: https://medium.com/@alanscottencinas/why-ai-orchestration-beats-bigger-context-windows-f65179b28854 Comments URL: https://news.ycombinator.com/item?id=48848606 Points: 2 # Comments: 0

A rewrite done in 11 days that would have taken a small team a year to complete, for $165K in tokens. Also: coding LLM “wars” heat up, AI fakers from North Korea still a problem when hiring, and more

Article URL: https://unskilled.blog/posts/are-we-governed-by-ai-yet/ Comments URL: https://news.ycombinator.com/item?id=48848561 Points: 1 # Comments: 0

A measured look at distributed training, from DDP and FSDP to the ZeRO stages in between, and why the wiring between your GPUs matters as much as the strategy you choose The post Behind the Scenes of Distributed Training and Why Your GPU Wiring Matters as Much as Your Strategy appeared first on Towards Data Science .

Article URL: https://github.com/riktar/memledger Comments URL: https://news.ycombinator.com/item?id=48848460 Points: 1 # Comments: 0

<p><strong><a href="https://ai.meta.com/blog/introducing-muse-spark-meta-model-api/">Introducing Muse Spark 1.1</a></strong></p> Following <a href="https://simonwillison.net/2026/Apr/8/muse-spark/">Muse Spark in April</a>, here's Muse Spark 1.1 - the first Spark model to offer an API. Meta claim significant improvements in agentic tool calling and computer use.</p> <p>There are a lot more details are in the <a href="https://ai.meta.com/static-resource/muse-spark-1-1-evaluation-report">Muse Spark

DesignOps is shifting from governing human-led process to orchestrating human and AI technology towards ethical, meaningful outcomes. Continue reading on UX Collective »

Article URL: https://ai-2040.com/ Comments URL: https://news.ycombinator.com/item?id=48848425 Points: 23 # Comments: 5

Article URL: https://twmrg.substack.com/p/the-quiet-overlap-how-personalized Comments URL: https://news.ycombinator.com/item?id=48848422 Points: 1 # Comments: 0

<p><strong>Release:</strong> <a href="https://github.com/simonw/llm-meta-ai/releases/tag/0.1">llm-meta-ai 0.1</a></p> <p>Let's LLM run prompts against the new <a href="https://ai.meta.com/blog/introducing-muse-spark-meta-model-api/">muse-spark-1.1</a> model.</p> <p>Tags: <a href="https://simonwillison.net/tags/llm">llm</a>, <a href="https://simonwillison.net/tags/meta">meta</a></p>

Article URL: https://marketplace.visualstudio.com/items?itemName=Codium.codium Comments URL: https://news.ycombinator.com/item?id=48848246 Points: 1 # Comments: 0

<p>Centralized Git was always going to break under agent load. The protocol that Linus shipped in 2005 was designed for a handful of humans running <code>git push</code> from a coffee shop — not a swarm of coding agents each running shallow-clone → edit → push in a tight loop, thousands of times an hour, against the same repository. By 2026, that constraint is showing up as GitHub rate-limit emails, flaking CI, and "the agent pool is throttled" Slack messages. The fix has been talked about for y

Article URL: https://vixsound.com Comments URL: https://news.ycombinator.com/item?id=48848202 Points: 1 # Comments: 0

<p><strong>Release:</strong> <a href="https://github.com/simonw/llm/releases/tag/0.31.1">llm 0.31.1</a></p> <blockquote> <ul> <li>Fix for a bug with OpenAI Chat Completion endpoints where a tool call with empty arguments could result in a JSON error from some providers. <a href="https://github.com/simonw/llm/issues/1521">#1521</a></li> </ul> </blockquote> <p>This bug came up when I was testing <a href="https://github.com/simonw/llm-meta-ai">llm-meta-ai</a>.</p> <p>Tags: <a href="https://simonwil

The soft, weirdly sexualized home-chore robot has been given some very tactile hands.

Humanoid robots at UC San Diego just successfully completed two surgical procedures.

<p>Last week I open sourced <a href="https://github.com/ronak-create/FableCut" rel="noopener noreferrer">FableCut</a>,<br> a Premiere-style video editor that runs in the browser and that AI agents can<br> operate. It hit the front page of Hacker News<br> (<a href="https://news.ycombinator.com/item?id=48845422" rel="noopener noreferrer">thread</a>), and the questions<br> there made me realize the interesting part isn't the editor. It's one design<br> decision: <strong>the project file is the inte

<p>TLDR; I got tired of babysitting N terminal tabs of five different coding-agent CLIs. So I built agentproto — one daemon that drives Claude Code, Codex, Hermes, opencode, and Mastra through the same lifecycle, and actually supervises them.</p> <h1> Why I built a daemon to drive every AI coding agent from one interface </h1> <p>I have a confession: at any given moment I have Claude Code, Codex, and<br> Hermes running in parallel terminal tabs, and I cannot remember which flag<br> spawns which,

<pre style='white-space:pre-wrap;width:81ex'>fix(agents): isolated cron busts prompt prefix cache via per-run session id (#96686) * fix(agents): isolated cron busts prompt prefix cache via per-run session id Isolated cron runs carry a per-run :run:<id> session scope (#91685) rendered verbatim into the cached system-prompt Runtime line, re-busting byte-exact prefix caching for the tool catalog after it every run (#96677, #43148 class). buildRuntimeLine now renders the stable base session key and

Article URL: https://www.pangram.com/blog/ai-in-your-feed Comments URL: https://news.ycombinator.com/item?id=48847940 Points: 84 # Comments: 62

MIT researchers developed FloatForm, a swarm of small aquatic robots that snap together like ants forming a raft, assembling into reconfigurable structures on the water.

Article URL: https://github.com/MadsLorentzen/ai-job-search Comments URL: https://news.ycombinator.com/item?id=48847930 Points: 2 # Comments: 0