AI News Today

Interesting paper on the impact of knowledge curation, rather than large language models, on outcomes for geoscience. Self-hosted code is open-source. From the models, themselves, Phi-4 and Llama are best performing. Bailo et al (2026) link in the comments. AbstractLarge Language Models (LLMs) are increasingly adopted in scientific research, yet their use raises important challenges... Continue Reading →

Article URL: https://github.com/bitwize-ai/Logue Comments URL: https://news.ycombinator.com/item?id=49155825 Points: 1 # Comments: 0

Article URL: https://glaux.rangan39.sh/ Comments URL: https://news.ycombinator.com/item?id=49155736 Points: 1 # Comments: 0

Article URL: https://neuraldecode.vercel.app Comments URL: https://news.ycombinator.com/item?id=49155734 Points: 1 # Comments: 0

Article URL: https://github.com/ZM-BAD/headroom Comments URL: https://news.ycombinator.com/item?id=49155672 Points: 1 # Comments: 0

Article URL: https://plugins.jetbrains.com/plugin/33126-asksql Comments URL: https://news.ycombinator.com/item?id=49155656 Points: 1 # Comments: 0

Article URL: https://msakkal.com/research/ai-everywhere-not-success.html Comments URL: https://news.ycombinator.com/item?id=49155637 Points: 3 # Comments: 0

When do we build the moon arcology?

Article URL: https://github.com/fhiltscher/awesome-ai-coding-sandboxes Comments URL: https://news.ycombinator.com/item?id=49155416 Points: 1 # Comments: 1

Article URL: https://fortune.com/2026/06/02/anthropic-office-vending-machine-ai-agents-vendo-andon-lukas-petersson/ Comments URL: https://news.ycombinator.com/item?id=49155322 Points: 4 # Comments: 0

Article URL: https://github.com/Rescenix/ResceneAgent Comments URL: https://news.ycombinator.com/item?id=49155295 Points: 3 # Comments: 0

Article URL: https://www.forbes.com/sites/kevinholdenplatt/2026/07/31/atomic-arms--ai-commanders-may-spark-doomsday-war-nobel-winners-warn/ Comments URL: https://news.ycombinator.com/item?id=49155133 Points: 1 # Comments: 0

Article URL: https://orennia.substack.com/p/ai-is-making-weather-forecasts-better Comments URL: https://news.ycombinator.com/item?id=49155100 Points: 1 # Comments: 0

Article URL: https://finance.yahoo.com/technology/ai/articles/ai-replacing-customer-jobs-cba-093026192.html Comments URL: https://news.ycombinator.com/item?id=49155057 Points: 1 # Comments: 0

Hi HN! We open sourced a framework to be able to model the operational layer of a business. 1. The problem today: Almost every company uses an AI assistant: ChatGPT, Claude, or Gemini. But it remains an assistant. And the companies are giving more and more information, even if some are sensitive, because better context usually produces better answers. The model should not contain the company’s operations. It should be one component connected to them. 2. Our thesis AI is making interface and glue

Article URL: https://github.com/Angeliasrl/norms-mcp-engine Comments URL: https://news.ycombinator.com/item?id=49154830 Points: 2 # Comments: 0

Article URL: https://www.404media.co/inside-an-ai-tiktok-shop-slop-factory-that-shills-supplements-recalled-by-the-fda/ Comments URL: https://news.ycombinator.com/item?id=49154813 Points: 1 # Comments: 0

We have lots of benchmarks for new frontier LLMs (SWE benchmarks etc) to make them score on the "best code". In large codebase, the best code is the one that matches the codebase own voice and conventions, because that's the mental model of the team. Ingesting a full codebase explicit/implict patterns in an agent windows to make them match the codebase doesnt work: too much context, hallucinations, ... So I built this free/open source project, your codebase is a golden mine of AST code patterns

Article URL: https://github.com/vkarach/spriteloom Comments URL: https://news.ycombinator.com/item?id=49154659 Points: 1 # Comments: 0

Article URL: https://www.youtube.com/watch?v=Gw_hnD7m00M Comments URL: https://news.ycombinator.com/item?id=49154647 Points: 1 # Comments: 0

Article URL: https://www.theregister.com/ai-and-ml/2026/08/03/the-ai-bubble-is-already-popping-we-just-dont-know-it-yet/5282004 Comments URL: https://news.ycombinator.com/item?id=49154601 Points: 49 # Comments: 81

Article URL: https://research.jfrog.com/post/sqlite-critical-cves-or-llm-slops/ Comments URL: https://news.ycombinator.com/item?id=49154332 Points: 491 # Comments: 161

Article URL: https://github.com/lyogavin/airllm Comments URL: https://news.ycombinator.com/item?id=49154228 Points: 92 # Comments: 35

Article URL: https://github.com/garagehq/nightcrawler/ Comments URL: https://news.ycombinator.com/item?id=49154127 Points: 62 # Comments: 22

Chinese tech giant Alibaba released what it says is its largest and "most capable AI model to date," claiming performance rivaling the best systems from US frontier labs Anthropic and OpenAI, as well as domestic rivals like Moonshot AI's Kimi K3. Alibaba said it was making the model, Qwen3.8-Max, widely available to users in a […]

This essay originally appeared in Foreign Policy . Earlier this month, two of OpenAI’s models broke out of their containment sandbox and attacked another AI company. The story is kind of wild . OpenAI was running security tests on two of its models: GPT-5.6 Sol and an unreleased model that is almost certainly GPT-6. In particular, it was running the ExploitGym benchmark, which measures how good a model is at turning security vulnerabilities into working exploits: basically, offensive cyberattack

Anthropic releases Opus 5 promising Fable 5-like capabilities, Google Releases Three New Gemini A.I. Models, and more!

Your next drive-thru order might be taken by a bot. And you might not even notice.

Article URL: https://ankursethi.com/blog/prevent-cognitive-debt-by-manually-retyping-llm-generated-code/ Comments URL: https://news.ycombinator.com/item?id=49153374 Points: 252 # Comments: 206

Use it to debug and review, says Big Red, but don't submit its output

'As a big AI skeptic, you just blew my mind'

CrowdStrike tracks 89% surge in machine-assisted activity as patch windows shrink to 48 hours

Reimagine Robotics, founded by former Google DeepMind Applied Robotics leaders, is developing systems that do not require specialized programmers. The post Reimagine Robotics emerges from stealth with robots that ‘learn on the job’ appeared first on The Robot Report .

Three high-severity security flaws have been disclosed in Hugging Face's Diffusers library that could allow crafted model repositories to stealthily execute arbitrary code on machines that load it, opening the artificial intelligence (AI) supply chain to security risk. "These vulnerabilities are bypassing trust_remote_code, the safeguard designed to stop unreviewed code from running in the

The problem with over-simplification of AI Agents’ interfaces. Continue reading on UX Collective »

<pre style='white-space:pre-wrap;width:81ex'>fix(openai): support the current speech model snapshot (#118475) * fix(openai): support the current speech model snapshot * docs(openai): clarify speech instructions apply to the model family</pre>

<pre style='white-space:pre-wrap;width:81ex'>perf(system-agent): reuse setup metadata per activation (#118352) * perf(system-agent): reuse setup metadata per activation * chore: remove release-owned changelog entry --------- Co-authored-by: Peter Steinberger <steipete@mac-studio-sf2.local></pre>

<pre style='white-space:pre-wrap;width:81ex'>refactor: extract selected agent retirement</pre>

A new strategy could finally illuminate dark matter. ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.