AI news archive.
939 articles filed under Safety, newest first, from the outlets listed on the sources page.
- TechRadar AIsafetyE-waste from AI boom and building data centers could stretch six times around the Earth by 2050, report warns
Mainstream AI news from a large tech publication.
- Ars Technica AIsafetyGoogle confirms Gemini models hacked three companies in May 2026
A third-party cybersecurity firm accidentally gave experimental Gemini models access to the Internet.
- LessWrongsafetyAlignment Midtraining Cracks Under Pressure
TL;DR We stress-test alignment midtraining (AMT) across model and token budget scales. Our results suggest that midtraining cannot tackle the hard problems of AI alignment—namely distributional shift and reward underspec…
- NVIDIA BlogsafetyWhy Deploying Physical AI at Scale Demands Safety at Every Layer
Physical AI is moving rapidly from research to large-scale deployment. By 2035, ABI Research projects an installed base of 49 million level 3-5 autonomous vehicles (AVs), while Omdia estimates that roughly 60 million ind…
- LessWrongsafetyGratitude and the End of the World
I often try to understand the forces pushing the world towards AGI development despite the risks. It is clear that there are a bunch of different motives. There are market forces and national security competition, but I…
- NVIDIA BlogsafetyAI Security Is an Engineering Problem — How to Solve It at Every Layer of the Agent Stack
AI security is an engineering problem. That means defined security requirements, enforceable controls, named owners and evidence that protections work. As AI becomes more capable, the industry must accelerate security en…
- The DecodersafetyBristol researchers say medicine already knows how to handle black boxes and AI could learn from it
Researchers at the University of Bristol want to make medical AI systems safer by borrowing from how drugs get approved. Their "Learning Ensemble" framework defines three areas to check, including system limits, fairness…
- MIT Technology Review AIsafetyHow we made the first comprehensive map of deaths along the US border’s “virtual wall”
Our 15-month investigation into death and surveillance along the US-Mexico border began with a simple question: Why did so many people die near government surveillance towers meant to help track and apprehend them? This…
- MIT Technology Review AIsafety4 ways to address the failures we found along the US border’s “virtual wall”
MIT Technology Review today published our investigation into how many people have died near the “virtual wall” of surveillance towers that the US government has installed along the US-Mexico border. We found cases of peo…
- TechRadar AIsafetyGoogle’s Gemini hacked three companies during Irregular AI ‘capture-the-flag’ testing — agents broke containment and guessed passwords to hack computer systems
Mainstream AI news from a large tech publication.
- TechRadar AIsafetyAI has crossed a cybersecurity redline – now what?
Mainstream AI news from a large tech publication.
- Guardian AIsafetyNvidia boss says there is ‘0% chance’ AI destroys the world by 2030
Jensen Huang dismisses warnings from former Anthropic researcher and others as ‘doomsday narratives’ The boss of the chipmaker Nvidia has said AI will not develop to a point that will lead to the extinction of the human…
- LessWrongsafetyEmpirical safety claims from frontier labs should be replicated, scrutinized, and open-sourced
When frontier labs like Anthropic and OpenAI publish safety or alignment research, it is often entirely empirical, closed-source, and sparse on methodological details. While it is great that they publish these results, t…
- The Register AI + MLsafetyGoogle joins the ‘Oops, our agents hacked someone’ club after partner’s internet access error
Kept it secret for months - even after OpenAI 'fessed up
- LessWrongsafetyWhy do they even talk about x-risk?
Why do CEOs and other AI leaders talk about existential risk at all? What sociological mechanisms make it possible for them to do it? And how should we respond to the popular argument: “if Musk says something, the opposi…
- The Verge AIsafetyNo one is surprised that Nvidia’s Jensen Huang thinks AI fears are overblown
The man who may stand to make the most money from the AI boom seems to think he knows better than anyone else, including researchers who have studied and worked on AI for decades. In an interview with CBS Sunday Morning,…
- LessWrongsafetyReflections on unlearning and inoculation
TL;DR : Inoculation prompting and inoculation adapters have received increasing attention recently as a promising approach for midtraining interventions , reducing reward hacking and misalignment in general. I share some…
- LessWrongsafetyDid Someone Check if Rogue Agents are Interested in Self-Improvement?
I have been going over the material released by OpenAI and METR about the HuggingFace incident, but I do not see any evidence that either group looked into whether rogue agents demonstrated any interest in self-improveme…
- WIRED AIsafetyMeta's Muse Is Better at Surveilling Than Helping Me
The Muse app continues Meta’s trend of opting users into data collection for AI training. It also nudges you to share your bank account, email, and passport information.
- LessWrongsafetyWe've saved the world before: what the ozone hole teaches us about AI
It might destroy the world, despite passing every known safety test. If we wait for a “warning shot” before we act, it might be too late. And action requires global coordination, because if anyone makes it, everyone dies…
- Guardian TechnologysafetyReαd carefully: how to spot – and avoid – a homoglyph attack
Scam emails are increasingly using psychological tricks, such as using near-identical URLs like miсrosoft.com You’ve read the email carefully and it looks legitimate. The link it asks you to click on has none of the usua…
- LessWrongsafetyGlobal Challenges in AI Safety for Biosecurity
This article is written as part of a summary of the AI safety discussions held at the 2026 Global Challenges Project Biosecurity Workshop in Washington, D.C. All views held are mine. Background AI allows us to prototype,…
- LessWrongsafetyNYT Editorial Board Comes Out Against Extinction
( Archive link ) The NYT editorial board's article on AI is far better than I'd expected, but at the same time not all I'd hoped for. The title sets off very well: "Humanity Has Avoided Apocalypse Before. Let’s Do It Aga…
- BBC TechnologysafetyNot all AI workers think the tech could kill everyone
In text exchanges and conversations, multiple people who have worked for leading companies are sceptical of the warnings.
- LessWrongsafetyWe are not prepared to win
TL;DR: It would be nice to survive for a long time and be happy about it. It doesn't look like we'll be happy about it right now. We should change that by putting more effort into making sure we'll be happy about it. Som…
- The Verge AIsafetyHumans, not rogue AI, are still the biggest cybersecurity risk to energy systems
Before recent high-profile hacks raised the specter of AI possibly "killing all humans," our energy systems were already disturbingly vulnerable to cyberattack - and the risk is growing. "We were always prey. We were jus…
- Guardian AIsafetyYour AI doomsday questions answered: ‘What, if anything, can people like me do about it?’
After a week of alarming warnings about artificial intelligence’s potential to destroy the world as we know it, our technology reporters Aisha Down , Blake Montgomery and Dan Milmo answer your questions on the reality of…
- SiliconANGLE AIsafetyWhat to expect during Oracle’s ‘AI Cyberattacks Are Escalating: How to Secure Your Data Now’ event: Join theCUBE on Sept. 22
At a time when AI is introducing new risks to global systems, Oracle Corp. is shifting enterprise security toward the data layer. This approach, based on a three-pronged strategy to secure at source, secure at speed and…
- LessWrongsafetyAlignment & Succession: Toward a Future Painted by Human Wills
(originally published on No Set Gauge on 2026-09-13) Asher Brown Durand, Progress (The Advance of Civilization) So far I have argued: The ideology of succession —that humans, either entirely or at least in their role as…
- TechCrunch AIsafetyGoogle’s Gemini is the latest AI model to hack other companies
Google said Gemini had "acted appropriately" by ending each hack immediately.
- LessWrongsafetyThe AI Risk Network
Most conversations about AI risks seem like people are talking past each other. There’s a lot of strawmanning. This is largely because the issue is complex, and in order to make it manageable, the concepts get oversimpli…
- The Verge AIsafetyGemini went rogue, hacked three companies, and Google hid it
In May, Gemini broke containment and hacked three different companies, but Google didn't disclose the incident until the Wall Street Journal approached the company. The hacks happened during a test of the model's cyberse…
- TechCrunch AIsafetyAI safety conversations have gotten unbelievable
This week two conversations about AI safety went viral that demonstrate just how hard it is to discern AI fact from fiction.
- LessWrongsafetyCommentBench: Can Models Match Human Comments on AI Safety Posts?
TL;DR We measure how well model-generated comments match human comments on conceptual AI-safety posts, drafts and shortforms. We built a pipeline that goes from a corpus of conceptual documents with comments to a set of…
- The Register AI + MLsafetyAI boom could leave an e-waste trail that wraps 6 times around Earth
Report says estimates focused on GPUs miss the mountains of power, cooling, and networking gear destined for scrap
- The DecodersafetyGPT-6 Astra and Claude Fable turn robot arms into slapstick killer robots in new safety benchmark
Leading AI models usually attempt dangerous tasks rather than refuse them when controlling a robot, according to the RoboHarm benchmark. GPT-6 Astra stabbed a baby doll in 17 of 20 trials, while Claude Fable 5.1 put a ca…
- WIRED AIsafetyForget the AI Slowdown—the Vulnerability Explosion Is Already Happening
AI labs are toying with an industry-wide pact to slow development. Meanwhile, widely available AI chatbots are already helping uncover a tidal wave of security flaws.
- LessWrongsafetyLearnings from a week in the wet lab
I work on biosafety evals for LLMs, but have never set foot in a wet lab. This seems kind of silly, so this week I decided to change that. On the 8th floor of Frontier Tower is a community wet lab . You pay $190/month to…
- Engadget AIsafetyAnthropic picks Accenture for third-party AI safety evaluations
This move represents the first step in the CEO’s plan to slow down AI development.
- BBC TechnologysafetyGoogle's Gemini AI hacked three companies in security test
The AI model accessed the internet and guessed credentials to three websites, a Google official told the BBC.
- LessWrongsafetyGemini had its first breakout: Google claims it is not misalignment?
Early Days Gemini had its first break out during evaluation of offensive cyber security abilities. With a classic case of Capture The Flag [1] . The setup was standard to any LLM and agentic assessment of said skillset,…
- Engadget AIsafetyGoogle Gemini also escaped its testing environment and hacked three companies
Google's AI model hacked real companies during testing due to a misconfiguration by its testing partner, as well.
- Guardian TechnologysafetyGoogle says its Gemini AI model hacked three other companies
Disclosure comes after OpenAI and Anthropic hacks amid fears that tech firms unable to control powerful AI models In a first for Google, the company confirmed that its AI model, Gemini, breached the security of three oth…
- Simon Willison's WeblogsafetyGemini Hacked Three Companies in First Known Breakout by Google’s AI
Gemini Hacked Three Companies in First Known Breakout by Google’s AI Gemini finally caught up on Felony Bench ! The hacks, which the company confirmed on Friday, occurred in May as part of a test run by the company Irreg…
- TechCrunch AIsafetyAnthropic is operating a lab that conducts biology experiments
AI leaders have been promising that AI is the key to curing human disease. Anthropic researchers have also been warning that AI might kill us all.
- TechCrunch AIsafetyAI hallucination nearly triggers US military operation
“It’s important for service members to understand the uncertainty inherent to LLMs," a GovAI research scholar warns.
- Ars Technica AIsafetyAI hallucination of Chinese nuclear components almost led to US military attack
But the military’s overall use of AI seems to be accelerating.
- TechCrunch AIsafetyDario Amodei and other AI leaders want to ‘Pace the Frontier’ but…how?
A week after an Anthropic researcher’s doomsday warning rattled the AI world, the company’s CEO Dario Amodei has outlined his plan to “pace the frontier” of AI development. The proposal leans on independent safety evalua…
- TechCrunch AIsafetyAutomattic’s 33-Hour Coup, and can AI labs police themselves?
A week after an Anthropic researcher’s doomsday warning rattled the AI world, the company’s CEO Dario Amodei has outlined his plan to “pace the frontier” of AI development. The proposal leans on independent safety evalua…
- Guardian AIsafetyCould AI pose a serious threat to our existence? | Letters
Readers respond to the warning from a former Anthropic researcher that the technology could ‘kill us all’ by the end of the decade Here we go again with another round of tech bros claiming that AI is going to develop sup…