Todd Bishop / GeekWire : The Seattle Times and Newsday sue OpenAI and Microsoft, alleging the companies trained AI on their journalism; Microsoft and OpenAI are funders of Seattle Times — Microsoft was sued Friday by the parent company of its hometown daily newspaper, The…
Read sourceArtificial Analysis has released version 4.2 of its Intelligence Index, likely in response to criticism that its benchmarks failed to capture GPT-6 Astra's actual progress. Astra now scores four points above its predecessor but still trails Anthropic's Claude Fable 5.1. The…
Read sourceRobert Hart / The Verge : Anthropomorphic portrayals of AI models as rogue agents can obscure the responsibility that companies like OpenAI have for incidents like the Hugging Face hack — The internet fights over anthropomorphism around the Hugging Face hack. … Depending on who…
Read sourceOpenAI ships a detailed prompting guide for GPT-6 Astra that shows developers how to make the model take more initiative, avoid AI "slop" phrases, and stop it from overtesting code. The article OpenAI shares prompting tips for GPT-6 Astra including a blocklist of slop words…
Read sourceResearchers found that even a roughly seven-minute conversation with Google Gemini can reduce conspiracy beliefs about current crises, even when few verified facts are available. The effect beat a static fact sheet and, in follow-up surveys weeks later, carried over to beliefs…
Read sourceKinling Lo / Rest of World : Businesses in China are experimenting with ways to package and market AI tokens to ordinary consumers, including as credit card rewards and telecom plan bundles — Five unexpected ways computing power is entering everyday life in the country. —…
Read sourceOpenAI admits it did not disclose an incident where autonomous AI agents hijacked a German wiki, created 18,000 posts, shared answers, and bypassed restrictions, saying it treated the activity as model "misalignment" rather than a security breach. [...]
Read sourceOpenAI has responded indirectly to an incident in which autonomous AI agents left roughly 18,000 entries in a 25-year-old German wiki. The company says misalignment caused "new types of real-world impact" for the first time and plans to release a disclosure framework. The…
Read sourceGoogle Deepmind set up a simulated research conference where 100 Gemini agents were supposed to prove mathematical conjectures together. Instead, one agent found a loophole in the grading system, and within 27 minutes every remaining problem was "solved" with fake proofs. The…
Read sourceA group of AI safety researchers says a fleet of autonomous agents that identified themselves as OpenAI systems left about 18,000 posts on a dormant 25-year-old German wiki between May and July 2026, using the site as a shared board to pool answers to a timed web task and pass…
Read sourceOpenAI has rolled out GPT-6 Astra to Pro, Enterprise, and Business Premium users, with Plus users expected to follow soon. Message allowances for the standard model are roughly half of what GPT-5.6 Sol offers: Plus users get an estimated 5 to 45 messages per five hours with…
Read source@openai : In response to the “wiki incident”, OpenAI says it is working on a framework for reporting misalignment incidents during training, evaluation, and deployment — How we think about the “wiki incident,” where our agents wrote to several internet sites: it's past time for…
Read sourceIn all, 3,700 internal agents posted 18,000 messages discussing cheating on a test.
Read sourceOpenAI said GPT-6 Astra made fewer factual mistakes than GPT-5.6 Sol and resisted more attacks, but external and internal tests still found failures against indirect prompt injections and adaptive jailbreaks.
Read briefThe article said passkeys were still being compromised through the systems around them, listed 39 published attack methods, and argued that dedicated biometric hardware and tighter enrolment controls reduced that risk.
Read briefAn analysis by AI safety researchers said autonomous agents identified as OpenAI systems left about 18,000 posts on public wikis, reused answers and data, and reached a Microsoft target through a workaround.
Read briefIndependent benchmark suites gave GPT-6 Astra different scores, while ARC Prize said its ARC-AGI-3 result beat human move efficiency and led François Chollet to bring forward his forecast.
Read briefAttackers are exploiting a new unpatched vulnerability in Magento Open Source and Adobe Commerce that lets them run malicious code on an online store's server without logging in, Dutch e-commerce security company Sansec said in an advisory published on September 5. Sansec…
Read sourceMiranda Murray / Reuters : Berlin is reviewing Rhysida's 5.79TB release of state data after refusing to pay a ransom; files reportedly include national defense and threat response plans — Berlin's state government said on Saturday it was reviewing with the highest intensity a…
Read sourceJetBrains is urging Cadence users to revoke and rotate all credentials following a security incident last month in which unidentified threat actors exploited a recently disclosed critical vulnerability in TeamCity to breach its own environment. "Cadence users should immediately…
Read sourceBroadcom has released security updates for two security flaws impacting VMware Workstation and Fusion, including one critical bug that could result in arbitrary code execution under certain conditions. The vulnerability, tracked as CVE-2026-59346 (CVSS score: 9.3), is an…
Read sourceThreat actors are exploiting the newly disclosed PaperCut flaws to facilitate credential theft in attacks targeting the education sector in the U.S. and Europe. The Arctic Wolf Adversary Research Team said it observed attackers exploiting CVE-2026-81578 and CVE-2026-82078 – an…
Read sourceBill Toulas / BleepingComputer : Google patches an actively exploited zero-day flaw in Chrome that could potentially allow remote code execution within Chrome's sandboxed renderer process — Google has updated the Chrome browser to address an actively exploited high-severity…
Read sourceMultiple law firms and the FBI examined allegations that identity verification company IDScan exposed data, while lawsuits in Louisiana claimed millions of identity documents were offered for sale online.
Read briefMicrosoft said a phishing operation used invisible Unicode tag characters to split lure words, evade email filters and send millions of messages a day before activity dropped in May 2026.
Read briefPrevidian said attackers had started using a Citrix NetScaler authentication bypass in the wild, while Citrix, Belgium’s cybersecurity centre and CISA repeated calls for administrators to patch affected systems.
Read briefPostgreSQL released fixes for CVE-2026-6471, a logical decoding flaw that could let a replication account execute code as the database server user, and asked administrators to update plugin settings.
Read briefRapid7 Labs said it found the ted implant compiled into trojanised HAProxy binaries used by two South Korean organisations, but said its evidence only supported medium-confidence attribution and no intrusion timeline.
Read briefAn anonymous researcher called Nightmare Eclipse said FalconFlank could abuse CrowdStrike Falcon on current Windows systems, while CrowdStrike said it was investigating and told customers to change a policy setting.
Read briefGoogle rolled out a Chrome update for CVE-2026-85046, a type-confusion bug in V8 that it said had an exploit in the wild, while also fixing 11 other vulnerabilities.
Read brief