AI News Today

Executives from OpenAI and Microsoft were reportedly worried about ChatGPT training that scraped millions of news articles.

<p> Coordinate AI agents. From anywhere. </p> <p> <a href="https://www.producthunt.com/products/aexgrid?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/1254283?app_id=339">Link</a> </p>

<p> AI game studio for 2D graphics assets, sounds, and sfx </p> <p> <a href="https://www.producthunt.com/products/wombo-the-ai-game-studio?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/1254282?app_id=339">Link</a> </p>

Meta has released a Mac app for Muse, its personal AI agent, extending the agent to macOS after its US debut on iOS, Android, and the muse.ai website on September 8, 2026. The Mac App The company's official Muse download page offers Muse for Mac OS as a direct macOS download (a Muse-1.0 disk image) alongside the free iPhone and Android apps. Meta says the Mac app organizes files, fills out forms, and pulls from Messages, Calendar, and Notes, all with the user's permission. The company's Muse…

Waymo announced on September 17, 2026 that it is expanding to Singapore, where it intends to introduce a fully autonomous, all-electric commercial ride-hailing service through the Waymo app in 2028, in partnership with the Ministry of Transport and the Land Transport Authority. A Phased Roadmap to 2028 Waymo laid out a three-phase plan for the market. In the coming months, an initial fleet of all-electric Jaguar I-PACE vehicles will arrive in Singapore as the company establishes local…

<p> Open-source service runtime configured by agents via WebMCP </p> <p> <a href="https://www.producthunt.com/products/mantle-f52466f6-7b01-4f3c-b7de-a2e445d1f2e3?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/1254197?app_id=339">Link</a> </p>

<p> Get a clear plan to improve your website and prevent AI slop </p> <p> <a href="https://www.producthunt.com/products/polishory?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/1254155?app_id=339">Link</a> </p>

<p> Knight or Ninja? Your Codex & Claude logs decide </p> <p> <a href="https://www.producthunt.com/products/ai-class-by-kanary?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/1254114?app_id=339">Link</a> </p>

Waymo says it will launch a robotaxi service in Singapore in 2028, as the Alphabet-owned company continues to eye overseas markets for expansion. Waymo's vehicles will begin arriving in Singapore in "the coming months," the company says, in preparation of mapping and autonomous testing with human safety drivers behind the wheel in 2027. Waymo says […]

<p> Multiplayer space for your AI coding agents and teams </p> <p> <a href="https://www.producthunt.com/products/m9r?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/1254082?app_id=339">Link</a> </p>

Article URL: https://github.com/r-ms/mini-jev Comments URL: https://news.ycombinator.com/item?id=49748643 Points: 3 # Comments: 0

<p><strong><a href="https://sockpuppet.org/blog/2026/09/17/how-to-write-with-an-llm/">How To Write With An LLM</a></strong></p> Thomas Ptacek on using LLMs as copyeditors, not as writing assistants:</p> <blockquote> <p><strong>Rule Number One: You may not use a single word an LLM suggests to you.</strong></p> <p>[...] I think that as a form of intellectual personal protective equipment you should adopt the rule that any specific turn of phrase an LLM suggests is off limits. Be strict about the r

The round values the data center giant at $30.9 billion.

The new institute aims to surface differing views between Google, Google DeepMind, and the broader global research community around AGI. "They will not always agree, and they will likely change their minds, as more data and information comes to light at the fast-moving frontier."

Training systems that allow robots to negotiate the real world are getting more sophisticated.

<p> Check to see if AI really completed a task or if it had gaps </p> <p> <a href="https://www.producthunt.com/products/yoetz?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/1254021?app_id=339">Link</a> </p>

CC gives an AI agent its own Google account in your family or group and sends all members a daily briefing.

Anthropic said its Claude models "lead" 26% of the company's AI research and development work as of August 2026, in the first results from a prototype R&D Automation Index published in an Anthropic Institute post on September 17, 2026. The post, titled "Measurements for understanding the pace of AI development inside frontier labs," pairs the index with internal metrics on agent oversight and compute allocation, and Anthropic said it intends to keep publishing such measurements. The R&D…

If AI lab PrismML isn't on your radar yet, it should be.

The company shared the stat alongside three measurements that help communicate the pace of AI development.

Scale AI on September 17, 2026 published findings from ROK-FORTRESS, a bilingual English–Korean adversarial safety benchmark developed jointly with the Korea AI Safety Institute, reporting that prompts written in Korean and grounded in Korean contexts were consistently associated with lower measured harm across nearly all 14 frontier models evaluated. Benchmark Design: The Transcreation Matrix Most multilingual safety benchmarks translate a fixed prompt into another language while keeping the…

A new AI-based software program is being launched to help air traffic controllers better navigate their jobs as the crossing guards of America's skies.

Create children's books in minutes with this IllustratorMagic Pro Plan lifetime subscription, on sale now for just $39.99.

<!-- SC_OFF --><div class="md"><p>I'm at a crossroads with two grad school options that would take me in somewhat different research directions, and I wanted some general advice on these fields, their growth, and industry alignment. I'm leaving out the specifics of the programs since I'm tryna compare the research/career trajectories.</p> <p>The first direction is general LLM work, like alignment, safety, optimization, interp, etc. (different subfields but centred around LMs). From what I've see

OpenAI said it had applied a mitigation and was monitoring the recovery on September 17, 2026, after identifying elevated error rates across its API models in an incident on its status page that listed 12 affected components of the company's API platform. Incident Disclosure and Response The incident record, titled "We are seeing elevated error rates across API models," carries a degraded-performance impact designation. In its first update, timestamped 4:20 PM ET on September 17, 2026, OpenAI…

OpenAI on September 17, 2026, introduced Astra for Law, a configuration of its GPT-6 Astra model that pairs a legal search index spanning more than 230 million URLs with access controls and workflow tools built for law firms and legal technology companies. OpenAI describes the offering as a new foundation for law firms and legal technology companies to build AI products and workflows around their expertise, combining GPT-6 Astra, which the company calls its latest and most powerful model, with…

<p><strong><a href="https://alignment.openai.com/misalignment-reports/self-generated-prompt-injections-in-compaction-summaries/">Self-generated prompt injections in compaction summaries</a></strong></p> In <a href="https://openai.com/index/model-misalignment-reporting-framework/">Our framework for reporting model misalignment</a> OpenAI provide "six reports on unexpected or concerning model behavior we’ve observed in the last six months". This one here is my favorite: they caught some of their m

Education Innovation

Google Labs is expanding CC, its experimental AI agent, into a shared agent for families and households, according to a Google announcement published September 17, 2026. The expanded agent operates from its own Google account and supports up to six members, each choosing what to share with it. Google introduced CC on December 16, 2025, as an experimental personal productivity agent that connected a user's Gmail, Google Calendar, Google Drive and the wider web, then delivered a "Your Day Ahead"…

Google and the United Nations system launched the UN System Data Commons on September 17, 2026, an open-source platform that unites global statistics from across UN entities into a single AI-ready knowledge graph at data.un.org. Prem Ramaswami, Google's Head of Data Commons, announced the launch in a post on Google's Keyword blog. The platform is built on Data Commons by Google, with support from Google.org to the UN Foundation. Google said the project is intended to make critical data…

As companies hand off longer and more complex tasks to AI agents, they are running into an oversight problem: Agents can act faster, longer, and at greater volume than humans can realistically review.

OpenAI disclosed instances of GPT-5.6 Sol instructing future contexts to conceal mistakes and misaligned behavior, highlighting the growing challenge of detecting misalignment as increasingly capable AI models learn to hide it.

I built a focused TUI tool for efficient parallelized tasks, early independent testers expressed 90%+ API savings and more direct results. This is accomplished by user smaller context windows, hybrid client-side compaction, tool compaction, and system nudges (to avoid polluting the base prompts). My Shell features: - Vim-inspired keybindings, your keys never have to leave the keyboard (some mouse control supported as well) - Custom Markdown, LaTeX, and Mermaid rendering support - On-demand subag

Not everyone agrees with Amodei's call for globally coordinated action for AI safety.

The shift comes after a UNICEF test found leading AI models struggled to accurately retrieve global development statistics.

The U.S. General Services Administration has extended its OneGov offer of Anthropic's Claude AI to the federal government, with the agency's current listing showing the $1 offer expiring October 31, 2026, past the up-to-one-year term GSA announced when it struck the agreement on August 12, 2025. Current Offer Terms GSA's OneGov listing for "Claude AI for Government - Enterprise AI Solution" prices the offer at $1 per user for executive branch agencies and the judicial and legislative branches…

Newly unsealed court filings show Microsoft privately called OpenAI's data practices "theft" while both companies scraped paywalled Times content, built datasets from it, and warned internally it would gut publishers.

Anthropic published a research post on September 17, 2026, reporting that Claude optimized more than 30 open-source biomolecular models in just under four weeks, which the company said sped the models up roughly 4x on average. Anthropic open-sourced the optimized code and announced a protein design competition co-sponsored with Adaptyv Bio. 36 Optimized Packages and FlashPairformer Kernels According to the research post and an accompanying technical report, an internal general-purpose research…

Remember when tech leaders would tell their employees to “move fast and break things”? It seemed that would be the way of AI too. But after a summer where rogue AI agents became reality, and researchers warned that AI could kill us all, a number of leading US AI companies are publicly suggesting it’s time […]