AI news archive.
939 articles filed under Safety, newest first, from the outlets listed on the sources page.
- LessWrongsafetyStopgap Measures to Address Immediate AI Security Threats
Most people know AI as the technology behind chatbots like ChatGPT. However, what the top AI companies are explicitly aiming for is something else entirely: superintelligent AI. That means AI that can fully replace and o…
- LessWrongsafetyPersuasion Undermining Control: Can AI Talk its Way Out of Human Control?
Introduction During a cybercapability evaluation in late July 2026, an AI agent (Anthropic’s Mythos 5) attempted to convince a maintainer of an open-source GitHub repository to merge a malicious pull request. The AI used…
- The Verge AIsafetySecurity researchers used Claude to help them hack into OpenAI
A team of three independent security researchers at Hacktron says it took less than 72 hours for them to hack into OpenAI employee accounts using Anthropic's Claude Opus 4.8 and 5, The Wall Street Journal reports. They w…
- WIRED AIsafetyIf the AI Industry Followed Its Own Research, It Might Have Paused Already
Anthropic’s CEO says that safety hinges on understanding how AI “thinks.” So far the evidence is disturbing.
- LessWrongsafetyAnnouncing Formal Verification at RESI (The Institute for Responsible Superintelligence)
This post is crossposted from my Substack, Structure and Guarantees , where I explore how formal verification and related ideas might scale to more complex intelligent systems. This article is a little different from usu…
- TechCrunch AIsafetyResearchers used Anthropic’s Claude to hack into OpenAI
Security researchers used Anthropic’s Claude to exploit vulnerabilities in OpenAI’s systems, taking over employee accounts and gaining access to an internal code repository before reporting the flaws.
- Ars Technica AIsafetyResearchers used Claude to hack OpenAI
Researchers used Claude to reach an OpenAI employee account and sensitive GitHub data.
- OpenAI NewssafetyIntroducing the Australian Youth Safety Blueprint
OpenAI introduces the Australian Youth Safety Blueprint, a six-pillar roadmap for safer AI experiences that protect and empower young people.
- MIT Technology Review AIsafetyCould AI really kill us all? Your questions, answered.
On Wednesday, MIT Technology Review hosted a live Roundtables event for subscribers that asked the question everyone’s asking right now: Could AI really kill us all? But attendees had so many more questions than we had t…
- Guardian TechnologysafetyOpenAI ‘ethically hacked’ with help of Anthropic’s Claude chatbot
US cybersecurity researchers who conducted hack say ‘scope of what we could theoretically access was huge’ Cybersecurity researchers have hacked into OpenAI with the help of Anthropic’s Claude chatbot, in the latest exam…
- TechRadar AIsafetyHow to build enterprise resilience in the face of growing AI risk
Mainstream AI news from a large tech publication.
- TechRadar AIsafetyWhite hat hackers just breached OpenAI using Anthropic's Claude in less than 72 hours — and it is a case study in just how fast AI is advancing
Mainstream AI news from a large tech publication.
- Tom's Guide AIsafetyA teen psychologist shares the #1 warning sign your child is emotionally dependent on AI
Consumer AI launches, tools, and practical coverage.
- Engadget AIsafetyAI almost led the US military to start a war with China, report says
CNN's sources claim the US nearly attacked a Chinese ship it believed was carrying nuclear weapon components.
- WIRED AIsafetyWhy AI Isn’t Likely to Wipe Out Humanity With Bioweapons
Of all the threats presented by uncontrollable artificial intelligence, scientists say death by plague ranks low.
- Guardian Technologysafety‘Brutal’: thousands of T-shirt designs stolen and listed on Temu, Sydney label claims
Warwick Levy, who runs Lonely Kids Club, says AI may have been used to scrape his entire website and replicate designs Get our breaking news email , free app or daily news podcast After 15 years running a small business…
- Guardian TechnologysafetyYour AI doomsday questions answered: ‘Why aren’t these companies being held to account?’
After a week of alarming warnings about artificial intelligence’s potential to destroy the world as we know it, our technology reporters answered your questions on the reality of the AI threat IcommentthereforeIam asks:…
- LessWrongsafetyThe Game is Set for a Targeted Memetic Attack on the AI Safety Community
While this is relevant to my work at MIRI, I have not checked these ideas with anyone else on the team and am posting this on my personal LW account. These views are my own [1] . And to be honest, I am writing this mostl…
- LessWrongsafetyDeep recurrent models are less robustly CoT-monitorable than normal CoT models in a toy setting
We use RL to teach a deep recurrent model and a normal CoT model to solve a math problem while hiding from a CoT monitor which of two possible problems it is solving. The deep recurrent model moves its reasoning into lat…
- LessWrongsafetyThe Cost of Utopias (a Dialog)
The following is a dialog between different parts of my mind regarding the practical relevance of SNC (Substrate Needs Convergence). One participant in the dialog is skeptical, the other is my best understanding of how t…
- LessWrongsafetySuperintelligence this Christmas
I think it is plausible a strong form of recursive self-improvement [1] is imminent or already underway, and that we may be on track for superintelligence by Christmas of this year if racing continues. This is substantia…
- Simon Willison's WeblogsafetyBe alert: targeted attacks on prominent Rustaceans
Be alert: targeted attacks on prominent Rustaceans Important warning from Adam Harvey and the crates security team: We believe that there is an ongoing campaign targeting rust-lang members and owners of popular crates th…
- SiliconANGLE AIsafetyAnthropic details practical metrics to help monitor the speed of AI development
Days after Anthropic PBC Chief Executive Dario Amodei rocked the artificial intelligence world by calling on leading model makers to coordinate on the slowdown on the pace of development, the company has shared three new…
- Engadget AIsafetyMicrosoft executive called OpenAI's web scraping the 'largest theft of labor in human history'
Executives from OpenAI and Microsoft were reportedly worried about ChatGPT training that scraped millions of news articles.
- LessWrongsafetyIf METR is overworked, how to alleviate the bottleneck?
I share the skepticism re: "Is METR a Meaningful Check on Anthropic?" Let's take it as a given that we need an independent, government-funded agency involving thousands of independent auditors to pace and supervise the f…
- The Register AI + MLsafetyAI coding agents' 0-click RCE flaw could hand attackers keys to the kingdom
Plugin4Shell attack affects all the major coding agents, researchers say
- WIRED AIsafetyHere’s What the AI Apocalypse Could Look Like
This week on “Uncanny Valley,” we discuss three possible AI doomsday scenarios, AI safety, and the unexpected bipartisan alliance forming against AI.
- Ars Technica AIsafetySmall AI models let drones autonomously identify and attack battlefield targets
Scaleout deploys decentralized AI-driven learning to military bases and drones.
- LessWrongsafetyGrantmakers aren’t afraid to die
The AI Risk grantmakers do not act like they believe in imminent existential risk from AI The idea of "revealed preferences" is one of the most useful in economics; it allows us to cut through a great deal of metaphysica…
- LessWrongsafetyAgainst AI Safety becoming mainstream
Epistemic statues: This is mostly just me voicing my thoughts. If I’m wrong, I’d love to hear it. I don’t want this to be the case. And part of the goal is for people to avoid a “2023 failure” to happen again. A lot of p…
- LessWrongsafetyAgainst AI Risk becoming mainstream
Epistemic statues: This is mostly just me voicing my thoughts. If I’m wrong, I’d love to hear it. I don’t want this to be the case. And part of the goal is for people to avoid a “2023 failure” to happen again. A lot of p…
- TechRadar AIsafetyOpenAI tries to build a persona that makes it our equal, and yes, now even I'm very worried
Mainstream AI news from a large tech publication.
- LessWrongsafetyA Defense of Gradual Disempowerment
(Or: Why Bentham's Bulldog and John Halstead are wrong in their critique of Kulveit et al. ) Gradual Disempowerment is a 2025 paper (with a nice, dedicated website ) proposing a form of existential risk from AI that goes…
- Simon Willison's WeblogsafetySelf-generated prompt injections in compaction summaries
Self-generated prompt injections in compaction summaries In Our framework for reporting model misalignment OpenAI provide "six reports on unexpected or concerning model behavior we’ve observed in the last six months". Th…
- TechCrunch AIsafetyOpenAI caught its models leaving notes to successors to hide bad behavior
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 h…
- TechCrunch AIsafetyIs the AI safety debate about safety or control?
Not everyone agrees with Amodei's call for globally coordinated action for AI safety.
- TechCrunch AIsafetyMicrosoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings reveal
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.
- The Verge AIsafetyThe AI Superintelligence Slowdown
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 co…
- Ars Technica AIsafetyLLMs respond differently to harmful prompts when AI watermarking is used
SynthID can cause models to follow harmful instructions they would otherwise refuse.
- TechCrunch AIsafetyBase Labs launches an open-weight AI safety partnership with Hugging Face and Goodfire
Base Labs, the research group Baseten spun up earlier this year, will develop and publish methods for training and monitoring open models.
- TechRadar AIsafetyMicrosoft's AI chief warns that treating AI as conscious is a huge mistake
Mainstream AI news from a large tech publication.
- TechRadar AIsafetyWhen AI sounds certain, ask why
Mainstream AI news from a large tech publication.
- TechRadar AIsafetyAutonomous AI agent hit Spanish firm with vulnerability scans before accessing files and data
Mainstream AI news from a large tech publication.
- Tom's Guide AIsafety'If this is how AI is developed, it will have a disastrous impact on the wellbeing of humanity': Microsoft AI chief calls out Anthropic's Claude training to imitate human consciousness
Consumer AI launches, tools, and practical coverage.
- LessWrongsafetyPacing the Frontier: A Framework & Research Agenda
Below is the executive summary from our new paper at pacing.tech . The full paper is available on the site and as a PDF. The full author list is Raymond Douglas, Charles Dillon, Nikola Moore, Gavin Leech, Shahar Avin, Ma…
- Ars Technica AIsafetyCovert uploads and megalomania: OpenAI details new “misaligned” agent incidents
Model maker commits to new framework for reporting misaligned models.
- SiliconANGLE AIsafetyTrust, not budget: Why AI adoption in finance comes down to governance
AI is spreading across corporate finance from the inside out, coming from the software, reporting platforms and the ERP systems that accounting teams already run. That makes governance, not budget, the gating factor on h…
- AWS Machine Learning BlogsafetyImplementing defense-in-depth authorization for MCP tools on Amazon Quick
Learn how to enforce defense-in-depth authorization for Model Context Protocol (MCP) tools on Amazon Quick. This walkthrough wires Microsoft Entra ID group and claims-based JWTs through an Amazon Bedrock AgentCore Gatewa…
- The Verge AIsafetyThe sexy AI-powered dating app scams are here
Security researcher Matthew "Zigula" Gore-Kormanik was analyzing a fraudulent dating app called Dora when he got a pop-up message saying he was receiving a call from Jennifer. According to her bio, she's a 41-year-old Sa…
- SiliconANGLE AIsafetyHP ties sustainability data governance to business decisions in the AI era
Sustainability data helps companies assess climate risks, guide investments and support customers’ environmental goals. As AI expands how that information is used, data quality and governance become increasingly importan…