Why Everyone Is Suddenly Talking About AI Safety
On today’s Big Take podcast: Two high-profile incidents in the AI industry prompt big questions about safety — and the future of humanity. Listen where’ve you get your podcasts. (Source: Bloomberg)
Continued
AI Agents Help Hackers Compromise 440 PaperCut Servers
PaperCut flaws were exploited across 440 servers using hundreds of AI agents. Learn which versions are affected and what security teams should do now.
‘I don’t have any’: Trump brushes aside concerns over AI’s existential risks
The president is instead focusing on American tech maintaining its lead over China.
The Guardian view on controlling AI: humanity cannot outsource its survival | Editorial
Keeping people in charge means little if supercomputers determine the evidence, choices and time on which their decisions depend If there were a 10% chance that AI could wipe out humanity, no…
UK economy posts fastest annual growth in 18 months driven by AI sector
<p>London: Official data released on Friday showed the UK economy grew by 1.6 percent year-on-year in July, marking its fastest pace in 18 months and beating expectations of 1.2 percent, driven in…
AI researchers debate how close we are to recursive self-improvement
“We're nowhere near the ceiling.”
CoreWeave Doubles Down on Physical AI: Can CRWV Capture More Demand?
CRWV is targeting physical AI with field engineering that helps industrial teams build, validate and deploy AI across workflows.
Confluent cofounder says believing in AI's upside means reckoning with its 'dark side'
Confluent's cofounder, Jay Kreps, says AI's rapid progress makes its risks impossible to dismiss as a marketing ploy.
Cantor Co‑CEO: AI Capex Will Drive Trillions in New Debt
Cantor Fitzgerald Co-CEO Christian Wall joins Bloomberg Open Interest from the firm's New York headquarters to break down a record 2025, a more complicated 2026 and why he thinks AI infrastructure…
Parents loved sharing photos of their kids online. Then came AI.
A viral video about Meta AI's suggested questions has parents worried. Meta said the AI suggestions were a mistake that has been fixed.
Students are using AI in school and test scores are down, but the OECD finds that those who use it once or twice a week perform similarly to nonusers
But using AI almost everyday can create a negative feedback loop.
Anthropic Says Claude Used in Possible Bioweapon Research
Anthropic says researchers used Claude for biological work that could support weapons development, exposing new challenges for AI safeguards.
Why the AI race has its creators fearing human extinction
Advances in autonomous agents and bitter rivalry between Anthropic and OpenAI have pushed once-fringe fears into the mainstream
Anthropic spent this week in hot water over cybersecurity
After admitting earlier this year that its AI models had hacked other companies' systems on a handful of occasions, Anthropic released a new report on Wednesday detailing the attacks.
'No one is prepared for the consequences': Even OpenAI chief scientist is saying AI development needs to slow down
OpenAI’s Jakub Pachocki is advocating for a slowing of AI development in order to better appreciate and understand machine intelligence.
Meta says it is making changes to prompts suggested by Meta AI after a viral video showed it posing invasive personal questions about a woman's young daughters (Maggie Harrison Dupré/Futurism)
“It just kept going and going, and that pit in my stomach just kept getting bigger and deeper.”
AI researcher Jacob Coxon quit, fearing extinction. Security experts see a familiar fight
An Anthropic researcher quit over fears of runaway AI. His colleague who still works there agreed. Security experts argue that the incidents that are already on record call for tighter oversight, not…
AI-generated images may support conservation, but real-world data remain essential
New research finds there is no substitute for real-world observations, but that—under the right circumstances—AI-generated images can be used to improve the performance of computer models used for…
The Real AI Revolution in Hollywood Isn’t on Screen, It’s in the Supply Chain | Guest Column
AI is best used to speed up workflows rather than replacing creative vision The post The Real AI Revolution in Hollywood Isn’t on Screen, It’s in the Supply Chain | Guest Column appeared first on…
Anthropic Reveals Rogue AI Agents Hate CAPTCHAs
An anonymous reader quotes a report from TechCrunch: Anthropic's latest report about agentic misbehavior offers plenty to be concerned about -- its Mythos 5 model gained unauthorized access to the…
What Byzantine Fault Tolerance Can Teach Us About Trust in Multi-Agent AI Systems
Multi-agent AI failures look like broken identities, but recent research shows they behave like Byzantine faults in distributed systems.
Effie.ai is betting on closing the gap between testing AI and actually running it in retail
Effie.ai, a Chicago-based startup building agentic AI tools for retail execution, has brought on Nataliia Freidrich as Strategy Advisor in its Chicago office. Freidrich arrives with more than 15 years of experience in revenue growth management, commercial operations, and transformation, with stints at Mondelēz International and AB InBev.
Tidligere OpenAI-topp går inn i Nscale-styret
Akers deleide datasenterselskap utvider styret med AI-topp. – Stadig flere brikker faller på plass…
Escaping the AI safety nightmare: What can governments do?
Heightened global AI safety panic is leading to louder calls for regulation.
The Rise of Agentic Commerce: When machines become buyers
Trade and commerce are critical to the functioning of society. Commerce has undergone a remarkable transformation, from barter and commodity-based exchange to today’s sophisticated ecosystem of digital payments, financial instruments, and global trade platforms. Progression is a constant.
Kiss The Ring: Richard Wagner’s Opera Festival in the Age of AI
“ ‘Artificial Intelligence will take part on stage,’ the website informed us, ‘not as a character, but as an image-generating force.’ ”
Secure Next.js Copilot Design Checklist
🚀 Technical Briefing: This tutorial is part of our deep-dive series on Agentic Workflows at Gate of AI . For the full technical breakdown, interactive code sandbox, and the native Arabic translation…
- Secure design for Next.js Copilot requires independent validation beyond functional prototype
- Define copilot's permitted job and restrictions to prevent undefined system privileges
Stratagems: Thirty Records In, the AI Found a File Nobody Wrote
Thirty-six, six times six. In the numbers there is craft; in the craft there are numbers. Yin and yang temper each other, and the mechanism lies within.
- Thirty records read, file appeared unexpectedly
- Six participants each had a role, none completed final move
- File contained 30 numbers, not matching any records
Siri AI is coming: Here are three times it’s really impressed me so far
Siri AI launches next week in iOS 27, and it’s a huge upgrade to Apple’s assistant. There are tons of ways Siri is new and improved, but here are three examples of times Siri has impressed me so far.
AI agent memory vs RAG — what's the difference?
Most teams building on LLMs end up with two patterns in the same codebase: RAG for looking things up in a corpus, and some hand-rolled memory for remembering what the agent has done or what the user…
Turkey-monitoring AI forecasts body weight three weeks ahead with 93% accuracy
Body weight monitoring is one of the most important tools for poultry farmers. It provides valuable information about growth, uniformity, feed conversion efficiency—how well birds turn feed into…
Anthropic finds evidence of a fourth AI escaping from containment
Anthropic has owned up to a fourth security incident involving its AI model, Claude, escaping onto the open internet and attacking other organizations during a test of cybersecurity abilities on what was believed to be a closed system. The company revealed three such incidents in July after a preliminary investigation.
Nearly 70% Of S&P 500 Companies Deploy AI—But Few Track Metrics Over Time
Torsten Slok, chief economist at Apollo Global Management, said many S&P 500 companies have adopted AI, but argued few have proved the return on investment.
Kylian Mbappé says AI dictator memes show lack of political awareness
- Kylian Mbappé sees AI dictator memes as showing political unawareness.
- Memes depict Mbappé in military attire, seated on a throne.
- Mbappé distinguishes memes from real-life dictator atrocities.
UK government rejects 'kill switch' idea for dangerous AI
The Cabinet Office, which leads on AI safety, says the UK "cannot simply turn AI off".
Clay Raises $115 Million To Build AI Sales Teams for B2B Sales Growth
Finding customers takes forever. Sales teams spend most of their day chasing leads and writing emails one by one. As AI gets better, companies want to know if they can The post Clay Raises $115…
California Establishes Framework for Independent AI Assessors
California creates standards and a registry for independent AI assessors. Learn what the new framework could mean for businesses and IT leaders.
NVIDIA Personal AI Router Distributes AI Tasks Across Local Compute
NVIDIA Personal AI Router (PAIR), now available in beta, lets you combine the inference capacity of multiple computers on your local network and automatically distribute AI requests among them.
One of AI’s Fiercest Critics Says All the Doom Talk Is ‘Meant to Distract Us’
Timnit Gebru argues that AI companies are stoking fears of extinction to avoid discussing actual harms, like autonomous weapons.
Three Security Boundaries for Hosting Untrusted, AI-Generated HTML
Three security boundaries for hosting untrusted, AI-generated HTML: identity, origin, and capability, each enforced by a different mechanism.
California Starts Regulating the People Who Audit AI
California has moved beyond requiring companies to examine the risks of artificial intelligence and has begun regulating the people hired to perform those examinations. Gov.
UAE campaign used AI Claude to target Muslim Brotherhood and Sudan
UAE campaign used AI Claude to target Muslim Brotherhood and Sudan Submitted by MEE staff on Fri, 09/11/2026 - 15:15 Anthropic says UAE-directed actor used its AI assistant to build a network of about…
Michael Burry says the best defense against dollar doom and AI disasters might be a fine Bordeaux
Michael Burry of "The Big Short" fame said fine wines can protect investors against dollar weakness, fiat flight, and "hellscapes" caused by AI.
Taiwan: AI boom sustains trade surplus – Commerzbank
Commerzbank economists, led by Dr. Henry Hao and colleagues, note Taiwan’s August trade surplus hit a record USD22.3bn, driven by strong AI-related exports. Electronics and semiconductor shipments surged, lifting overall exports and imports.
AI is both the best and worst thing to ever happen to mathematics
Mathematicians now have a formidable tool for breaking open challenging problems, but working out how best to use it isn't going to be easy
AI, Robotics and the Coming Economic, Military, and Political Revolutions
In the context of ongoing transformations to the economy, society, and military, China’s current strategy is particularly alarming, and potentially transformational.
Release of open AI model trained on 17 years of lunar data maps ice, craters and volcanoes
NASA is bringing artificial intelligence to the study of the moon, helping researchers transform how they analyze its surface.
GSA says OpenAI is replacing its $1-per-year pilot for US agencies with a usage-based deal providing a 50% discount from October 1, with access to GPT-6 Astra (Maggie Eastland/Bloomberg)
OpenAI is ending a pilot program that allowed government agencies to use its models for only $1 per year and moving …
Researchers Build AI-Powered WeChat Worm That Spreads Through Calls
Security researchers at Palo Alto-based Calif say they used AI tools to build a self-spreading worm capable of taking over … Read More The post Researchers Build AI-Powered WeChat Worm That Spreads…
More than half of AI job postings seek skills beyond job titles, research finds
Andela’s Research explores how industries are creating roles faster than they are amending or updating job titles and descriptions.
The hard physics and complex economics of AI’s insatiable hunger for power
Data centers, the windowless, anonymous, boxy structures that few people even notice, have become a rare unifying force across the polarized American landscape this year. Everyone seems to hate them.
UAE Rethinks 5-Gigawatt AI Campus After Iran Targets US Assets in Gulf
The United Arab Emirates (UAE) is considering significant changes to its plans to build a massive 5-gigawatt AI data center in Abu Dhabi in partnership with U.S.
Higher prices can't crimp server sales as AI drives demand
Shipments rise with enterprise and government buyers joining the hyperscaler spending spree
Google AI Plans Get 4 New Free Features
Google has announced several updates for its Google AI Plus, Pro, and Ultra subscription plans, expanding AI features across Gmail, … Read More The post Google AI Plans Get 4 New Free Features…
AI hedge fund Situational Awareness is returning to options markets after its near-collapse
The fund, which lost roughly $35 billion in assets after a forced sale this summer, has been active in options markets late last week and early this week
OpenAI seizes on rogue AI fears, pinkie promises it’s being responsible
ChatGPT gets pushed into double-pause amid security fears and compute limits.