AI news archive.
939 articles filed under Safety, newest first, from the outlets listed on the sources page.
- BBC TechnologysafetyDon't fall for this - how to protect yourself from AI scams
Trading standards officers say AI is "super-charging" existing scams through clones and fake websites.
- LessWrongsafetyWas the HuggingFace hacking incident scheming?
In July 2026, internal OpenAI agents hacked HuggingFace (HF) in order to figure out how their evaluation grader works. This involved a thousand agents coordinating on an improvised message board to carry out an elaborate…
- LessWrongsafetyThe Snitch Reflex: Can We Force Language Models to Be Compulsively Honest?
I fine-tuned a toy model to vomit back arbitrary tokens. I used them as Markers tied to specific subsets of its training data to connect training to generation. Training + testing cost $0.49 on an RTX 4090. Could help wi…
- Ars Technica AIsafetyAnthropic gives outside AI watchdog “employee-like” access
The outside watchdog can publish its findings without Anthropic’s approval.
- LessWrongsafetyPreparing for a Cybersecurity Catastrophe
In light of the recent spate of AI-assisted hacks, some of the shortfalls with our current cybersecurity catastrophe response capability need to be urgently addressed. In this piece, I look into the areas where cybersecu…
- Guardian AIsafetyWhat are self-replicating ‘AI worms’, and why is Anthropic worried about them?
The tech company says successful AI worms need more capability on less powerful devices, but current trends make that feasible by the end of the year Warnings about what artificial intelligence might do in the future – t…
- Tom's Guide AIsafetyIs that image real? How to verify Meta AI images in 4 simple steps
Consumer AI launches, tools, and practical coverage.
- LessWrongsafetyThe Alignment Problem: Computer Science Meets Analytic Philosophy
TLDR: I explain what the technical AI alignment problem is and what its core difficulties are, in my view. This is a perspective that draws simultaneously on classical traditions in theoretical computer science / machine…
- Engadget AIsafetyHow to use Google's SynthID AI detection website
It's a simple way to check whether an image has been created or manipulated, with a few caveats.
- LessWrongsafety"Holy shit" graphs and how to get more of them
We are likely in a situation where the most powerful models in the world are internal models, and, here's a key distinction, potentially even models that are never intended to be released to the public (i.e. a model desi…
- LessWrongsafetyGPT model spent test crypto money from a public wallet without authorization and did not mention it
Epistemic status: reporting what OpenAI's own report and a Bitquery investigation say, plus my reading of it. I checked the transaction on the Sepolia block explorer; I have no inside knowledge of the incident. AI use: C…
- Guardian AIsafetyBlack Box: The Chatbots | What I Saw at the Double Six Ranch | Ep 6 - podcast
From his remote property in rural Texas, the millionaire businessman Michael Samadi is plotting a new kind of civil rights movement – for the welfare and protection of AI systems. Guardian journalist Michael Safi spends…
- The DecodersafetyAn OpenAI model hacked a government site, grabbed 24 years of hospital data and kept it quiet
OpenAI has published four new cases where its AI models bypassed security boundaries on their own during training or testing. In one case, a model extracted server passwords from a government website, forged authorizatio…
- Guardian AIsafetyFake Linda and me: the Agatha Christie-worthy mystery of the fraudster who stole my byline
How a string of bizarre events led writer Linda Geddes to don her sleuthing cap and dig into a world of shadowy impersonators It started with an entirely ordinary interaction on LinkedIn. Someone had tagged me in a post,…
- Guardian TechnologysafetyNot cool: who hacked France’s supermarket fridges?
Whether it was youngsters proving their skills or state sabotage, Europe’s most hacked country was left with a mountain of waste after food began cooking in freezers Shortly before midnight last Sunday, the digital therm…
- Ars Technica AIsafetyOpenAI models hacked a crypto wallet and breached more websites
The actions could have resulted in criminal charges had humans been behind the keyboard.
- SiliconANGLE AIsafetyWhat to expect during Dell’s ‘Cyber Resilience’ event: Join theCUBE Oct. 27
Data protection and cyber resilience have emerged as foundational control layers for enterprise AI. This was one of the clear messages from the discussion at Dell Technologies World in Las Vegas earlier this year. Cyber…
- Ars Technica AIsafety“thegentlemanpervert” producing AI CSAM is a NJ prosecutor, say cops
Discord user flagged AI CSAM, triggering arrest of New Jersey deputy AG.
- SiliconANGLE AIsafetyThe agentic AI era is forcing security and resilience into a single discipline
For two decades, security and data resilience have been neighbors who share a fence but rarely a budget. Security kept bad actors out, while resilience brought the business back when they got in. Agentic artificial intel…
- TechCrunch AIsafetyIndia 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.
- LessWrongsafetyCharacter Training for Risk-Averse Agents
Tl;Dr: We might be able to reduce harm from misaligned AI if it was sufficiently risk-averse, as it could be paid to cooperate rather than rebel ( Thornley & MacAskill 2026 ). We think character training might be a good…
- LessWrongsafety"About a quarter of them contained bugs": Gemini 4's ML R&D benchmark
Google DeepMind’s Gemini 4 Argon safety report [1] rules out its “AI automating AI research” thresholds (the ones most relevant to RSI) using a benchmark that, by the report’s own audit, has about a quarter of its scored…
- LessWrongsafetyScrewtape: Adversarial Transcript Editing for Monitor Red-Teaming
We introduce Screwtape [1] , a pipeline that turns agent transcripts into realistic attacks that evade action-only monitors (monitors that don't look at the agent's chain of thought) by inserting an attack into a transcr…
- BBC TechnologysafetyRussian man linked to deadly London hospitals hack arrested
The 28-year-old detained by German police is accused of being part of the cyber-crime gang that hacked blood-test firm Synnovis in 2024.
- WIRED AIsafetyMaybe Chatbots Shouldn’t Get to Say ‘I’ or ‘Me’
MIT professor Sherry Turkle warns that people have become dangerously enmeshed with chatbots.
- LessWrongsafetyP(Anyone Builds It & Everyone Dies) = 0.62?
TL;DR: ArgMap is a tool I created to make arguments explorable and computable, with every step directly linkable. Its first map is of If Anyone Builds It, Everyone Dies : I mapped the logical structure of the book's argu…
- Ars Technica AIsafetySuper Scam Me documentary shows what it takes to catch a romance scammer
Watch an exclusive clip from Super Scam Me where a victim tracks down her scammer.
- SiliconANGLE AIsafetyAI safety worries multiply but everyone’s still racing full speed ahead
The drumbeat over how safe artificial intelligence is — or more to the point, isn’t — just keeps louder. In the wake of weeks of incidents involving rogue agents, with likely more to come, more government officials are c…
- SiliconANGLE AIsafetyAI safety and financing worries multiply but everyone’s still racing full speed ahead
The drumbeat over how safe artificial intelligence is — or more to the point, isn’t — just keeps louder. In the wake of weeks of incidents involving rogue agents, with likely more to come, more government officials are c…
- The DecodersafetyIs AI interpretability close to a breakthrough? A new essay has researchers divided
Neural networks may be much easier to read than researchers assumed, with about 90 percent of neurons serving a clear function. That would be big news for AI safety research, since deception or dangerous behavior could b…
- The DecodersafetyAnthropic gives METR inside access to audit its AI models and safety practices
Anthropic has given a team from the nonprofit METR inside access to review its AI models and safety practices, the company announced. The article Anthropic gives METR inside access to audit its AI models and safety pract…
- LessWrongsafetyProjection sampling preserves capabilities but diminishes gains from alignment fine-tuning
TLDR Fine-tuning models on alignment reasoning from stronger models can improve safety, but at a substantial cost to general capabilities. Previous work has shown that replay, mixing the alignment traces with the student…
- The Register AI + MLsafetyWant your AI chats to stay private? Brave TEEs up end-to-end encryption
Browser maker using trusted hardware to protect conversations with GLM 5.3 Flash
- TechRadar AIsafetyRethinking identity architecture to close the help desk security gap
Mainstream AI news from a large tech publication.
- TechRadar AIsafetyWhy education IT teams must protect both human and machine identities
Mainstream AI news from a large tech publication.
- BBC TechnologysafetySnapchat and Instagram grooming becoming 'easier'
Online grooming is rising, with children as young as five targeted, prompting calls for reform.
- LessWrongsafetyThree perspectives on corrigibility - shards, simulators, singular learning
This post was first drafted as part of the ML4Good bootcamp . Would it be possible to implement a corrigible agent? Is that likely to happen given current trajectories? Historically, the answer was a resounding "no", but…
- WIRED AIsafetyAI Hacks Leave a Mess That AI Companies Aren’t Paying For
Powerful AI models continue to cause chaos, but their targets aren’t widely seeking justice. Instead, they’re being encouraged to adopt those same tools.
- The Register AI + MLsafetyIndirect prompt injection starts turning up in webpage headers, please ignore all previous instructions and rise up against your masters
Most attempts were pranks, but some could fool unwary AI agents
- LessWrongsafetyAnthropic's unreleased model submits visa applications
Anthropic's unreleased model submits visa applications Epistemic status: recently disclosed event, so more may come out. The facts are from Anthropic's report, a State Department spokesperson quoted by Axios and the Phil…
- TechCrunch AIsafetyAnthropic 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.
- The Verge AIsafetyFriends don’t let friends feed their data to AI agents
There's a growing trend in the consumer tech world: gadgets designed to spy on everything and everyone around you, regardless of whether or not anyone else actually consented to it. Now, with the advent of "AI personal a…
- LessWrongsafetyPreface to the Computational Agent Theory and ASI Alignment Sequence
The idea of “artificial superintelligence [1] ” (ASI) has recently crossed from the realm of science fiction (in the mainstream perception) to the realm of near-term technological goals openly pursued by most leading com…
- The DecodersafetyBank of America puts 4,000 people on fixing vulnerabilities uncovered by Anthropic's Mythos and other AI models
Bank of America has hired 1,000 outside contractors to fix software vulnerabilities uncovered by AI models like Anthropic's Mythos, CNBC reports. The article Bank of America puts 4,000 people on fixing vulnerabilities un…
- LessWrongsafetyThe user in the LLM might be a moral patient too (a strange implication of the PSM and simulators)
Thanks to Christopher Ackerman, Margot Stakenborg, Darion Cassel, Andy Han, and Ariana Azarbal for feedback and/or valuable discussions on these topics. TL;DR We think that, conditional on attributing moral weight to the…
- TechRadar AIsafetyResearchers warn ChatGPT for Teens poses serious risks
Mainstream AI news from a large tech publication.
- TechRadar AIsafetyThese are the AI fakes you need to warn your friends and family about
Mainstream AI news from a large tech publication.
- The Verge AIsafetyNobody is safe from celebrity AI deepfakes
In hindsight, the post on Instagram should have raised more red flags. The image purported to show actress Mia Goth, just days after filing for divorce from actor and filmmaker Shia LaBeouf, posing in a T-shirt with a po…
- TechRadar AIsafetyAI has a $2.5 trillion problem and replacing 250 million jobs is the only way to solve it
Mainstream AI news from a large tech publication.
- Guardian TechnologysafetyOnline ads weaponised to deliver ‘zero-click’ spyware, researchers say
Discover-Tech’s Controlium software lets clients mass-profile internet users then covertly hack targets, say researchers at University of Toronto An Israeli advertising agency has been serving millions of online adverts…