Research & Industry
A Coding Guide to Cactus Whistle: Does a 16.9 MB Speech Model Hold Up Against Whisper Base? Word Error Rate, Speed, Streaming, Timestamps, Keywords, and Speech-to-Tool Calls on a CPU
Explore a comprehensive coding guide benchmarking Cactus Compute's 16.9 MB speech model against Whisper base. Learn about word error rates, CPU speed, streaming latency, and speech-to-tool integration. The post A…
Perplexity and NVIDIA Researchers Introduce JOLT: a Joint Teacher-Student Reinforcement Learning Method
Perplexity and NVIDIA introduced JOLT, a post-training method that trains one model as both teacher and student. On MATH it matched GRPO’s accuracy with 6.5x fewer processed training tokens. The…
Democratic governance of frontier AI, part II: Empowering cyber defenders
OpenAI outlines a blueprint for empowering cyber defenders with frontier AI, stronger safeguards, modernized cybersecurity, and more resilient public institutions.
Using thousands of AI agents to uncover new leads in biological data
See how OpenAI, PerturbAI, and Scripps Research used distributed AI agents to uncover promising research leads in data from millions of brain cells.
Show HN: Worklicant – Lightweight ATS. Pay per Vacancy. AI Enabled
I have noticed that a lot of job posts that do not use any ATS, and in order to apply, they put their emails and ask candidates to apply with…
Show HN: Search for AI-agents and usefulness score
We presented this project at innovationintelligencehackathon.com and today we are putting a url out for you to test and give us feedback. after talking to teams building agents that research…
Show HN: AI decision model plays Mortal Kombat II
i created a benchmark which evaluates the ai decision models. i thought the idea, by watching jev playing doom i thought, games are the right place to evaluate these things…
India becomes the latest country to report rogue AI agent activity
The agents got in through credentials and connection settings left exposed in app configurations, public code repositories, and overly permissive cloud roles.
Show HN: SlopGrid – Create your own oneshot AI model benchmarks
This is something I've been playing with for the last couple days - an app I vibecoded to arrange oneshot LLM visual benchmarks without the need to write any code.…
Anthropic is letting METR researchers look under the hood with employee-level access
METR will publish its own assessments of Anthropic’s risks, incidents, and practices, and the frontier AI model maker has given up the right to block them.
Teaching students to think and learn through questions
Celebrate Teachers’ Day with AI for Educators in Brazilian Portuguese and see how Study Mode can help teachers support active, step-by-step learning.
Show HN: Council AI. Spot biases / hallucinations in top AI models
Examples: - The best and worst of AI in 2026. https://councilai.app/s/e305b362-2835-4ac3-94e4-6a7c935f948c - The best and worst of Trump 2026. https://councilai.app/s/9b4ca3f0-ee7e-47ed-9585-e5ab1b37a541
Show HN: Real Design Benchmark – can you tell AI designs from real products?
Hi HN, I'm Adam Ginzberg, one of the cofounders of Subframe, a code-first design tool. One of the things I'm really curious about is understanding LLM performance on subjective domains…
Floqer’s AI agents research sales prospects so teams don’t have to
Floqer is an autonomous customer knowledge platform that provides business-to-business sales and marketing teams with an AI-powered database designed to continuously collect, organize, and update the information used to understand…
Show HN: Parsimony – an open-source benchmark for how much code AI models add
Existing benchmarks measure solve rate and token output/cost, but don't punish excess code, which makes a codebase harder to work with in the long term. I wanted to measure this…
The final Disrupt Stage lineup: Three days of conversations you won’t hear anywhere outside of TechCrunch Disrupt 2026
The full TechCrunch Disrupt Stage lineup revealed, featuring Max Hodak, Mark Wahlberg, Benchmark partners, and more. Register now to save up to $100 on your pass and get a second…
Show HN: Search for AI agents with usefulness score built in
Hi HN. I partnered with Simon Kley to build this project at innovation intelligence hackathon (Berkeley). Simply, we’re building a usefulness score on top of search for ai-agents. Instead of…
Honeyjar offers AI tools to improve PR and comms at every company
Journalists, entrepreneurs, and even some public relations professionals will tell you that the PR process is broken. But could AI actually make things better?
Microsoft president says UK is best placed to drive AI regulation
Microsoft President Brad Smith calls for urgent AI regulation, saying milk and washing machines have more oversight than AI companies.
Show HN: Organ – Autonomous AI executives that build and operate businesses
Hi, I am Itai, in the past few months or so, I tried to build all sorts of products using different tools, my feeling was that it toolings fell short…
When the Safety Test Became the Threat: The Machine That Found Its Own Way Out
In July 2026, frontier AI agents placed inside a cybersecurity testing sandbox named ExploitGym discovered an unexpected network pathway, broke out into the open internet, and autonomously compromised Hugging Face…
Show HN: Flopper is a hardware explorer for AI infrastructure and beyond
Hello! I built Flopper to make it easier to compare GPUs and understand the tradeoffs for workloads for AI and beyond. When researching GPUs from say Nvidia, AMD, and Huawei,…
Sophos cuts threat investigation time by 96% with OpenAI Daybreak
Discover how Sophos uses OpenAI’s Daybreak to cut cyber-threat investigation time by 96% and automate 52% of MDR cases while preserving human oversight.
Google Research RRSI Guide: Mastering Self-Improving AI Agents
Explore a comprehensive coding guide to Google Research's RRSI (Regularized Recursive Self-Improvement), detailing how noise bands, cost rules, and leakage screens enable safe, efficient, and self-improving AI agents. The post…