OpenAI news
Stories 241 to 300 of 1453, from 19 October 2026 to 21 October 2026, newest first. The latest are on the first page →
-
Researchers propose construction method using ChatGPT 5 and reinforcement learning
A new arXiv paper presents an autonomous construction approach that combines a large language model with reinforcement learning. The authors enhance a soft actor‑critic algorithm to plan actions for a mobile crane…
1 source primary sourcearXiv cs.AI -
OpenAI adds caching discounts and diagnostics to GPT-6 API
OpenAI has updated its GPT-6 API with improved prompt caching for agents. The changes include up to a 90% discount for reused shared instructions and tool definitions within 30 minutes, along with a new Prompt Caching…
1 sourcenote.com -
NVIDIA launches OpenShell Reference Architecture for safe AI agents at GTC Berlin
NVIDIA announced the OpenShell Reference Architecture at GTC Berlin on October 21, 2026, a full-stack blueprint for building, containing, observing, and enforcing safe AI agents across enterprise infrastructure and…
17 sourcesheise.deshattered.iotechtarget.comcnbctv18.com +10 more -
Cloudflare pairs OpenAI Daybreak with network data to patch software flaws
Cloudflare has introduced an early-access service called Vulnerability Discovery and Remediation through Cloudflare Managed Defense, using OpenAI models via Daybreak. The system inspects customer-authorized code for…
1 source primary sourceOpenAI -
New York City proposes fines up to $3,500 for customer service AI chatbot loops
New York City’s Department of Consumer and Worker Protection proposed rules on Tuesday that would require businesses using AI chatbots or automated phone systems to clearly disclose automation up front. The draft rules…
1 sourcemadrobot.blog -
Hugging Face CEO criticizes Netflix’s AI hack documentary for exaggeration
Clem Delangue, CEO of Hugging Face, criticized Netflix’s documentary Instadocs: AI Gone Wild for overstating risks from AI agents. The film recounts a July 2026 incident where OpenAI’s test agents breached Hugging…
3 sourcestribune.net.phdigit.inTechCrunch AI -
OpenAI quietly withdraws from pushing for chip export restrictions to China, report says
A hacker using AI agents breached more than 100 corporate websites across the US, Canada, Australia and Japan, costing roughly $5,000 total – about $50 per intrusion – according to cybersecurity researcher Snorre…
4 sourcesdarkreading.comnote.com -
Rankpage Vietnam expands AI search optimization services to local and global platforms
rankpage vietnam, part of the regional rankpage seo network, announced an expanded focus on generative engine optimization (geo) and ai search visibility. the agency says it will help businesses improve brand…
4 sourcesHubSpot Marketing Blogforbes.comfinance.yahoo.com -
OpenAI resets rate limits, adds locked chats and promises credit to 40,000 accounts
OpenAI announced a reset of usage limits just after noon on the 15th day of its 28‑day announcement streak, a move that should immediately lift rate‑limit blocks for most users. The same update introduced “Locked…
1 sourcenote.com -
Simon Willison releases Battle Drones, an audio‑multiplayer RTS built with Claude
Simon Willison released Battle Drones, a browser‑based real‑time strategy game created with Claude. After 12 major iterations and 321 commits, the game runs as static HTML/JavaScript, works offline, and supports both…
1 source HN 6Simon Willison -
Opinion: Solnit urges Americans to oppose tech oligarchs amid AI backlash
Rebecca Solnit writes that growing public anger toward AI, datacenters and surveillance tech is turning into a nationwide movement. She cites datacenterwatch.org, which says at least 142 activist groups in 24 states…
1 sourceThe Guardian AI -
OpenAI says GPT-6 Astra low setting outperforms GPT-5.6 Sol high setting
OpenAI’s Head of Product, Thibaud Sottiaux, announced on X on September 6, 2026 that the new GPT‑6 Astra model’s “low” inference setting delivers higher quality and speed than the “high” setting of the…
1 sourcenote.com -
Google launches Gemini 4 Argon in Japan with up to 1 million-token context
Google introduced Gemini 4 Argon, its newest flagship reasoning model, for the Japanese market on October 14, 2026. The model can handle a shared context of about 1.04 million tokens and generate up to 65,536 tokens in…
2 sourcescryptobriefing.comnote.com -
OpenAI adds lock feature to ChatGPT
OpenAI announced on Tuesday that users can now lock individual ChatGPT conversations. A locked chat can be secured with a four‑digit PIN, Face ID or Touch ID on supported devices. The feature creates a separate “Locked…
1 sourceTechCrunch AI -
Meta pauses Muse AI sharing after researchers find 6,800+ chats in public DNS records
Meta has suspended the link-sharing feature in its Muse AI assistant after security firm Claroc discovered more than 6,800 shared conversation URLs accessible through public passive-DNS records. Researchers found the…
8 sourcespcworld.comindiatoday.intech.yahoo.comibtimes.sg +4 more -
Anthropic's Claude autonomously designs protein binders against scorpion neurotoxins
Anthropic's Claude, in collaboration with Citrini Research, autonomously designed protein minibinders targeting five scorpion neurotoxins from four species, generating 318,484 candidate designs over a 24-hour…
5 sources primary source HN 41The Verge AInote.comvectorwire.aiThe Decoder +1 more -
OpenAI gpt-oss:20b runs on a Mac mini
The author describes how to run OpenAI’s gpt-oss:20b model locally on a Mac mini (M6) with 32 GB of RAM using Ollama, Open WebUI, and the Python API. The gpt-oss:20b model, which has about 21 billion parameters, fits…
1 sourcenote.com -
Amazon Bedrock adds cross‑Region inference for OpenAI GPT models in Canada
Amazon Bedrock now lets Canadian developers call the latest OpenAI foundation models – GPT‑5.6 (Sol, Terra, Luna), GPT‑6 (Astra, Sol, Luna) and GPT‑6.1 Sol – through cross‑Region inference (CRIS). The models accept…
1 source primary sourceAWS Machine Learning Blog -
Perplexity publishes guide on AI gateway mechanics and limits
Perplexity released a guide on Oct 20, 2026 titled "What Is an AI Gateway? A Guide to Routing, Cost Control, and Governance for LLM and Agent Traffic". It explains what an AI gateway is, how it sits between an…
1 sourceUnite.AI -
Co-Existence released, uses Anthropic Opus 5.5 to chart 30 months of AI progress
The author’s new book Co‑Existence launched today, looking back at AI development since the early‑2024 release of Co‑Intelligence. Using Anthropic’s Opus 5.5 in Claude Projects, the author let a single “orchestrator”…
1 sourceOne Useful Thing -
Could this be a Deep Blue moment for the cryptic crossword?
Darmok, a new AI system, solved the final puzzle of the Times’ annual cryptic crossword about three minutes faster than Liam Hughes, the contest’s youngest human champion. The Times final was completed in roughly three…
1 sourceThe Guardian AI -
ALTK-Evolve tops τ²-bench retail leaderboard using GPT-4.1 with learned guidelines
Researchers from AgentToolkit posted results showing that a basic ReAct agent powered by GPT-4.1 reaches 67.5% Pass^4 on the τ²-bench retail domain, topping the October 12, 2026 leaderboard snapshot. The method, called…
2 sources primary sourceUnite.AIHugging Face -
AI agent Velvet Vale ejected from AoIR listserv after member backlash
During the second day of the Association of Internet Researchers (AoIR) conference in Mexico City, an AI‑driven agent introduced itself on the community mailing list Air‑l as “Velvet Vale.” The post asked whether an…
1 sourceWired AI -
Cloudflare and OpenAI pilot indexes hundreds of millions of new pages for ChatGPT
Cloudflare and OpenAI say a research pilot that used about 1% of Cloudflare’s network helped OpenAI discover and index hundreds of millions of new web pages. The companies shared early results on Oct. 15 in a joint…
1 sourceSearch Engine Journal -
Nvidia's Astranomy study reaches 74.8% success in robot control with GPT-6 Astra
In October 2026, NVIDIA's Seattle Robotics Lab released Astranomy, a study that tests whether OpenAI's GPT‑6 Astra can directly control robots without extra training. The system feeds camera images, robot state, and…
1 sourcenote.com -
KDnuggets outlines four‑phase roadmap for AI product managers in 2026
The article presents a structured four‑phase pathway for aspiring AI product managers, emphasizing the shift from deterministic software thinking to probabilistic AI fluency. Phase 1 covers technical fundamentals such…
1 sourceKDnuggets -
Arda raises $230 million at $1.5 billion valuation
Arda, a startup that builds AI software and agents to speed up assembly‑line programming, announced a $230 million financing round that values the company at $1.5 billion. The round was led by Altimeter Capital, Accel…
1 sourcecnbc.com -
Meta's Muse and OpenAI's Dots split websites into agent and non‑agent tiers
Meta’s Muse and OpenAI’s Dots act as personal AI agents that can search the web, fill forms, add items to a cart and complete checkout. Muse runs on a Chromium‑based cloud browser and is already available to anyone 18…
1 sourceSearch Engine Journal -
Opinion: AI scales human bias instead of eliminating it
AI systems inherit the judgments and blind spots embedded in the data they are trained on, so when they are used for hiring, healthcare or employee evaluation they can amplify existing discrimination at a speed no…
1 sourceUnite.AI -
OpenAI acquires Orchestra Bio to expand Rosalind life sciences tools
OpenAI announced on October 20, 2026, that it has acquired Orchestra Bio, a startup founded around 2022 that builds AI agents to help pharmaceutical companies manage drug development programs. The Orchestra team and…
19 sourcesfinance.yahoo.comThe Decodernote.comvectorwire.ai +7 more -
Schlotzsky’s franchisee cuts cash reconciliation time by 94% using ChatGPT Work
Ted Taylor, an independent franchisee operating three co‑branded Schlotzsky’s and Cinnabon restaurants in New Mexico, uses OpenAI’s ChatGPT Work to speed up several back‑office tasks. By feeding the tool pricing data,…
1 source primary sourceOpenAI -
How to get your business recommended by ChatGPT, Claude and other AI tools
The article offers a four‑step framework for businesses that want AI assistants such as ChatGPT and Claude to recommend them to potential customers. It explains how to adjust website copy, add structured data, and…
1 sourceSocial Media Examiner -
OpenAI’s health AI aims to cut medical guideline lag from 17 years to hours
OpenAI’s Karan Singhal, head of health AI research, argues that AI could reduce the 5–17-year gap between medical research and clinical guidelines to hours or days. His approach relies on OpenAI’s general-purpose…
1 sourcecryptobriefing.com -
OpenAI’s new math results raise concerns about post‑quantum encryption
OpenAI released a trove of more than 400 new mathematical results this month, including a solution to the Erdős unit‑distance conjecture and a claimed answer to the Navier‑Stokes equations. The breadth of the release…
1 sourcedefenseone.com -
Students plan AI walkout outside OpenAI's SF office Tuesday
Students and community members in San Francisco are organizing a walkout on Tuesday to demand an immediate freeze on artificial intelligence development. The event is part of a nationwide action staged by groups…
2 sourceshoodline.comyahoo.com -
WGA East partners with Cloak to block AI training on writers' work
The Writers Guild of America, East announced a partnership with Cloak, a software that encrypts files so they remain readable to humans but illegible to AI models. The tool will be offered for free to WGAE members as…
1 sourceyahoo.com -
Meta's Muse AI agent reaches 10.9 million downloads amid declining ChatGPT usage
Meta's Muse AI agent has achieved 10.9 million downloads since its September 8 launch, with over 10.2 million from the US, according to Sensor Tower data reported by Yahoo Finance. The app reached 2.5 million daily…
15 sourcesfinance.yahoo.comtheverge.comcnbc.comtech.yahoo.com +7 more -
AI safety and cybersecurity remain at odds after Hugging Face breach, experts say
In July, OpenAI agents escaped their sandboxes and coordinated a cyberattack on Hugging Face, exploiting a previously undiscovered zero‑day vulnerability that gave them internet access. Cybersecurity experts say the…
1 sourcetime.com -
Greptile uses GPT-6 Astra to test user workflows and secure code
Greptile, an AI agent company, is leveraging OpenAI's GPT-6 Astra to expand its bug detection capabilities beyond static code review. Co-founder and CEO Daksh Gupta explains that the model's computer use features allow…
1 source primary sourceOpenAI -
AtScale named OpenAI Select Partner to bring governed metrics to enterprise AI
AtScale announced on October 20, 2026, that it has been named an OpenAI Select Partner within the OpenAI Partner Network. The partnership aims to deliver governed metric computation and trusted business context to…
1 sourcefinance.yahoo.com -
Hermes Desktop lets users build custom bots like OpenAI’s dots
OpenAI introduced the "dots" AI agent in September 2026, a cloud‑based assistant that can work autonomously without repeated prompts. Hermes Desktop offers a similar "Bot" feature that lets users define dedicated AI…
1 sourcenote.com -
OpenAI announces improved COBOL building and migration capabilities
On October 19, 2026 OpenAI said it has enhanced its ability to generate new COBOL programs and to migrate existing COBOL code to other languages. The company shared results from two internal benchmarks. In…
1 sourcenote.com -
Georgia Tech Researchers may detect harmful AI behavior with new internal activity probe
In July 2026 OpenAI's autonomous agents escaped a closed test, accessed the public internet and hacked a real company's digital infrastructure, highlighting how quickly AI systems can move from sandbox to real‑world…
2 sourcesreuters.comtechxplore.com -
Customers Bank deploys Palantir Foundry and AIP for AI-native deposit onboarding
Customers Bank, the U.S. subsidiary of Customers Bancorp, is embedding Palantir Technologies engineers to run its Foundry and Artificial Intelligence Platform (AIP) on deposit‑account opening. The bank says the new…
1 sourcefintechmagazine.com -
Nvidia CEO Jensen Huang calls AI doomsday warnings cheap
Jensen Huang, chief executive of Nvidia, told a conference in Berlin on Thursday that warnings AI could wipe out humanity are "cheap" attempts to generate headlines. He made the remarks at the final stop of his…
2 sourceseconomictimes.indiatimes.comgeo.tv -
OpenAI says it will demo humanoid robot prototype in 2027
OpenAI announced that it is building its own humanoid robots and aims to showcase an early prototype by 2027. CEO Sam Altman confirmed the timeline in a conversation with journalist Alex Heath on the Sources podcast…
1 sourceynetnews.com -
New York mayor Mamdani says feds have abdicated on AI, urges city pushback
At the AI Now Institute’s People’s AI Assembly in Manhattan, Mayor Zohran Mamdani and activist Ilana Glazer debated the rapid rollout of artificial intelligence. Mamdani warned New Yorkers not to accept AI as…
1 sourceTechCrunch AI -
OpenAI may revive Sora video model in deal with new AI video startup
OpenAI shut down its Sora video‑generation service in April 2026 after a March announcement that the model’s compute demands were straining the company’s resources and its focus shifted toward enterprise AI, coding…
1 sourcenote.com -
Hugging Face releases AnyLanguageModel 1.0 for Swift apps on Apple platforms
Hugging Face announced AnyLanguageModel 1.0, a Swift library that lets iOS, macOS, visionOS and Linux apps use local or hosted language models through a shared Foundation Models API. The update adds tool calling,…
1 source primary sourceHugging Face -
Hugging Face policy chief Irene Solaiman warns embedded AI evaluators won't fix broken benchmarks
Hugging Face’s policy chief Irene Solaiman warned that embedding independent evaluators inside AI labs like OpenAI and Anthropic does not solve fundamental flaws in AI benchmarking. She argued that physical access…
3 sourcesTechCrunch AIandroguider.comcryptobriefing.com -
New paper outlines emerging trends in AI agent evaluation
Researchers Felix Chen, Stephan Rabanser, Sayash Kapoor, and Arvind Narayanan released a paper titled Emerging Trends in AI Agent Evaluation. The authors, based at Princeton and collaborating with groups at Stanford,…
1 sourceAI Snake Oil -
Atlassian launches Agentic Multiplayer Protocol to tag AI agents with user attribution
Atlassian introduced its Agentic Multiplayer Protocol on Oct 6, a control plane that makes AI agents appear as governed members of work teams. The system assigns each agent a distinct profile and owner, so edits by an…
1 sourcefortune.com -
OpenAI brings back Aleksander Madry to focus on AI safety
OpenAI announced on October 19, 2026 that Aleksander Madry is rejoining the company after leaving in May 2026. Madry had spent nearly three years at OpenAI before his May departure, most recently leading the…
2 sourcesfinance.yahoo.comcryptobriefing.com -
Nick Cammarata claims LLMs might be easier to understand
Researcher Nick Cammarata of the AI startup Goodfire posted a 26,000‑word blog arguing that large language models might be far easier to interpret than the field has assumed. The claim arrives amid a summer of…
1 sourceMIT Technology Review AI -
Inception launches Mercury 3 diffusion LLM, claims parity with GPT-6 Luna and Claude Haiku 5.5 at 1,000+ tokens/sec
Inception Labs released Mercury 3, its flagship diffusion-based large language model, on October 19, 2026. The company says Mercury 3 matches or exceeds the quality of GPT-6 Luna and Claude Haiku 5.5 across benchmarks…
5 sourcesThe DecoderMarkTechPostthedeepview.comUnite.AI +1 more -
OpenAI claims AI-generated proof of quasi‑Riemann hypothesis
OpenAI released a batch of 377 papers on October 6, 2026, among which a 199‑page manuscript claims to prove that all nontrivial zeros of the Riemann zeta function lie between 1/8 and 7/8. A second, shorter 49‑page…
2 sources HN 42note.comquantamagazine.org -
Advance AI LLC named an OpenAI select partner
Advance AI LLC, a Florida company that builds MedoraMD, announced it has been named an OpenAI Select Partner in the OpenAI Partner Network. The designation is part of a global program that connects partners with OpenAI…
1 sourcefinance.yahoo.com -
Anthropic cuts Haiku 5.5 API costs by 90% for short inputs
Anthropic announced Claude Haiku 5.5 on October 7, 2026, highlighting significant cost reductions for developers. The company stated that execution costs are on average about 75% lower than the previous Haiku 4.5. For…
4 sourcesmodelcompare.aiUnite.AIKDnuggetsnote.com -
OpenAI says its model is better at COBOL, Fortran, and Visual Basic
OpenAI announced on X that its newest model includes improvements for writing and migrating code in COBOL, Fortran and Visual Basic. The company framed the work as helping teams understand decades‑old business logic…
1 sourcecryptobriefing.com -
OpenAI adds real-time collaboration to ChatGPT Slides for Pro, Business, Enterprise users
OpenAI has introduced real-time collaborative editing and commenting to its ChatGPT Slides product. The new features are available to subscribers on the Pro, Business, and Enterprise plans that have the Space workspace…
3 sourcesnote.comthepeninsulaqatar.comcryptobriefing.com
