AI News Today

SPONSORED FEATURE: While linking the GPUs in a datacenter is a challenge, coordinating thousands of them to work efficiently over hundreds of kilometers presents a new set of difficulties.

Our background is trading systems and exchanges. Liquid Inference applies that to LLM inference. On Liquid Inference, model providers compete with each other on an open marketplace to offer the best price for any given model. We paired that with an intelligent auto-router so you always pay the lowest cost on the spot for a preset quality level. Currently we're cheaper on Kimi/GPT/GLM but working hard to be the best price for all models. You can also customize your minimum routing settings if you

Local pharmacy phoneline is answered by "Caroline" with convincing voice, mannerisms and even background typing noise. You only realise she's fake at the first stochastic parrotese. My bank sends emailed query responses signed by e.g. Sarah. No disclaimer, e.g. "AI can make mistakes". Again starts convincing but can degenerate into parrotese. If I politely ask bot/human?, I get "human"... but if I query nonsensical advice, I get a non-resolving answer classic of ChatGPT. The reverse centauar in

Analysts welcome Fusion Claw's tighter controls but warn of early adoption and cloud lock-in

Without the ability to benchmark Large Language Models (LLMs), it is difficult for consumers and businesses to understand what progress a model has made over recent versions, and how it stands up to its competitors: Since LLMs are non-deterministic (i.e., they will not always produce consistent outputs given the same inputs), evaluating them is tricky. Even when researchers use identical prompts, model settings and benchmark datasets, seemingly minor differences in the execution environment can…

President Donald Trump has a knack for turning words against his enemies. His first successful presidential run was built on monikers like "Little Marco" and "Crooked Hillary"; he changed "fake news" from a phrase describing scammy media outlets to a derogatory term for the press at large. Over the past few weeks, he's clearly decided […]

<p> Review your code and your agents' changes before you ship </p> <p> <a href="https://www.producthunt.com/products/gitglow-2?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss-feed&utm_source=producthunt-atom-posts-feed">Discussion</a> | <a href="https://www.producthunt.com/r/p/1274980?app_id=339">Link</a> </p>

The cost of control The post Where Does the Money Go Across Long-Running Coding Agents? appeared first on Towards Data Science .

Before there were cute little guys, there was Instinct. In August, the startup got its AI agent to market with an unusual playbook: invite-only, no marketing, and barely so much as a website. And yet, Instinct quickly became the buzziest thing in AI, garnering praise for its straightforward, text message-based interface and its ability to […]

MCP just got simpler. Learn how to build a stateless MCP server in Python and expose tools, resources, and prompts over HTTP.

Humans have looked to the natural world for technological inspiration for millennia. AI and rapidly improving biotechnology are giving us new ways to explore four billion years of evolutionary experiments. The post We’ve Always Copied Nature. Now We Can Search It. appeared first on SingularityHub .

Buyers rushed to beat hikes earlier this year, but demand has since dried up as AI-fueled component shortages send costs soaring

<!-- SC_OFF --><div class="md"><p>ALHR - Adaptive Learnable Hierarchical Routing, uses static binary trees and learnable functions to minimize the amount of keys to be read.</p> <p>It does use a dense teacher while phase 1 of training however.</p> <p>MQAR TEST AT 1024 TOKENS - Average keys read per query by dense - 512 Keys</p> <p>Average keys read per query by ALHR - 30 keys</p> <p>Top - 1 accuracy of dense - 94.9%</p> <p>Top - 1 accuracy of ALHR - 92.1%</p> <p>KV Compression of dense - 1x(100%

<table> <tr><td> <a href="https://www.reddit.com/r/MachineLearning/comments/1x1lbrm/261008927_can_ai_agents_make_openended_scientific/"> <img src="https://external-preview.redd.it/q3evP6JeDpAC2MdSQHWYxnCYTqbJkElIQsLFqVSdkss.png?width=640&crop=smart&auto=webp&s=de730fbf7ecace6df0036b21470c16a2d4feacfb" alt="[2610.08927] Can AI Agents Make Open-Ended Scientific Discovery? Evidence from Station" title="[2610.08927] Can AI Agents Make Open-Ended Scientific Discovery? Evidence from Station" /> </a> <

It is kind of a huge deal.

(Martin Harvey/The Image Bank RF/Getty Images) Taking back their island home. ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

For years, cybersecurity has been obsessed with visibility. We wanted to see more of the network, collect more telemetry, correlate more events, and detect threats earlier. So we built dashboards. Then we built better dashboards. Then we added machine learning to those dashboards. Now we are adding AI. But at some point, we have to ask the obvious question: once we can see the attack, what are we actually prepared to do about it? I don't think the next big cybersecurity problem is going to be…

Anthropic on Thursday unveiled OSS Scanner as an opt-in vulnerability scanner to help secure the open-source ecosystem using artificial intelligence (AI). "It's an opt-in service informed by our experience using Claude to find vulnerabilities during Project Glasswing," Anthropic said. "Projects that join will receive thorough, periodic security scans by our strongest models at no cost."

Article URL: https://github.com/athrael-soju/Narwhal Comments URL: https://news.ycombinator.com/item?id=50019698 Points: 2 # Comments: 1

Shield AI matches the funding as autonomous X-BAT heads toward flight testing later this year

The firm said users can no longer engage in "sustained and needless" abusive behaviour towards the tech.

AI Agents for Beginners, Part 1 The post Build Your First AI Agent with One Tool Call appeared first on Towards Data Science .

Groceryshop 2026 not only featured mobile robots for taking inventory, but it also showed how retail automation is evolving. The post Groceryshop 2026 shows that retail robots are ready to scale, use AI appeared first on The Robot Report .

I loved the stunning 3D designs created with Astra but they typically require a decent knowledge of Blender or Unreal engine which made it inaccessible. It also generated something you couldn't edit which is tricky for the design process where you typically need to try different ideas out. This is especially painful if each series of Astra loop's took 4 hours to complete. So... using Arcadium's online house editor tool - I created a skill to generate JSON that it can understand. It's still early

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. We’re putting too much faith in AI’s ability to say no Today’s AI models are trained to refuse a vast number of prompts. If you ask your chatbot how to poison…

The bias-variance tradeoff describes tension between models that are too rigid to capture real structure and models that react too strongly to the particular training sample. This guide explains the mechanism, trade-offs, evaluation, and controls that matter in practice.

Build a production-grade data science toolkit without spending a dollar, using open-source AI tools that match, and sometimes exceed, their paid alternatives.

OpenAI fired three safety researchers who helped investigate the Hugging Face hack. In an open letter, they warn that the firings are scaring remaining staff and eroding safety culture. OpenAI claims they violated policies but won't say how. The article OpenAI's safety crisis keeps getting worse and the company keeps making it worse appeared first on The Decoder .

As enterprises race to deploy autonomous AI agents to accelerate business, a new report reveals they are tethered to security architectures built for a different era. The "Horizons of Identity Security" report from SailPoint highlights a critical “velocity paradox,” in which organizations invest in AI-speed business operations while continuing to rely on human-speed security controls, creating a

Explore Microsoft's Surface Laptop Ultra preview featuring local AI agents, Nvidia RTX Spark architecture, and Windows OS updates.

Anthropic is offering open-source projects a new way to check for vulnerabilities with its OSS Scanner.

Students are engaging with AI faster than schools can respond. Districts need a real literacy strategy, not just a usage policy. New York City gave 600,000 elementary and middle school students an answer to the AI question, and it's no. The nation's largest school district announced a one-year moratorium on student-facing generative AI through eighth grade, calling it the most “expansive ban of its kind in the country.” High schoolers get restricted access and mandatory "critical thinking"…

<p>Liability lawsuits have held tobacco, oil and pharma companies to account in the past. AI investors won’t ignore this threat</p><p>The most basic function of government is to protect people from harm. Two growing phenomena – the climate crisis and AI – pose escalating risks of <em>extraordinary</em> harm.</p><p>The climate crisis is already causing floods, wildfires, drought and record heat. AI agents are already escaping super-secure environments to hack into systems they’re supposed to avoi

AI agents can now write more code than any of us can read. That changed how I think about AI. The post Stop Using AI. Start Hiring It. appeared first on Towards Data Science .

Apple’s Mac desktops have become some of the most sought-after computers this year, driven by surging interest in agentic AI. Here’s which model to buy depending on your use case.

Article URL: https://github.com/SpectraSI/SpectraAdamw Comments URL: https://news.ycombinator.com/item?id=50018712 Points: 2 # Comments: 0

New usage policy cracks down on cruelty to chatbot, alongside slightly more pressing matters of weapons development, surveillance, and election interference

I tested AI-assisted interface generation before designing more than 800 interconnected enterprise screens. The test struggled with visual… Continue reading on UX Planet »

(Pai-Shih Lee/Moment/Getty Images) This explains a lot. ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

In its 26th season premiere, the procedural legal drama paints a damning picture of power-mad AI CEOs.