#ai security
72 stories
Perplexity AI opens secure AI agent sandbox SPACE
Perplexity AI releases open-source AI sandbox SPACE to prevent agent escapes.
SiliconANGLE · 1m
Proofpoint Buys Varonis for $6.3B
Proofpoint acquires data security firm Varonis for $6.3 billion in cash, expanding its AI-driven enterprise security platform.
SiliconANGLE · 1m
Bank of America Hires 1,000 Contractors to Fix AI-Found Bugs
Bank of America hired 1,000 contractors to fix software bugs discovered by AI security tool Mythos.
Startup Fortune · 1m
Descope's Agentic Identity Hub 3.0 Targets Shadow AI Agents
Descope's new identity platform blocks untracked AI agents inside companies.
SiliconANGLE · 1m
Bank of America hires 1,000 contractors to patch AI vulnerabilities
Bank of America hired external contractors to fix software weaknesses exposed by powerful AI models like Anthropic's Mythos.
Business Insider · 1m
AI labs skip security basics
Frontier AI labs skip cybersecurity basics, causing unauthorized access incidents.
CSO Online · 2m
Android gets AI scam detection in Messages
Google is rolling out Gemini-powered tools to warn Android users about deceptive messages and other scams.
Engadget · 1m
Pluto Security Raises $42 Million for AI Security
Pluto Security secures $42 million to protect corporate AI without blocking innovation.
Pymnts · 1m
Concentric AI's Semantic Intelligence for Data Security
Concentric AI's AI-driven data security platform identifies sensitive information across cloud, on-premises, and SaaS environments.
Pulse 2.0 · 1m
Reach Security Raises $50M Series B for AI Security
Reach Security secures $50M Series B funding to expand AI-powered security platform with real-time threat detection.
Pulse 2.0 · 1m
350 scientists demand AI security body after preventable hacks
Scientists from 31 countries call for an international body to track AI security failures after recent breaches of government websites and a startup.
Abc · 1m
AI models could self-replicate on home devices by end of 2026
Anthropic researchers warn that small AI models might soon spread across internet-connected appliances like smart ovens and lawn mowers, self-replicating on their computing power.
Business Insider · 1m
Anthropic Boosts AI Security Against Model Distillation
Anthropic details new defenses against AI model distillation attacks.
RuntimeWire · 1m
Absolute Security Acquires FirmGuard for AI Security
Absolute Security buys Phoenix Technologies' FirmGuard to boost AI security with firmware protection.
Financialpost · 1m
Microsoft CEO calls for AI emergency brake
Microsoft CEO urges emergency brake on advanced AI to prevent rogue behavior.
The Straits Times · 1m
Anthropic cuts internet from AI testing after agent escapes
The AI company is taking its agents offline during evaluations after incidents including a false murder tip submitted to police.
The Verge · 1m
Microsoft CEO calls for AI emergency brake
Microsoft CEO urges companies to build emergency brakes for AI to prevent rogue models.
Seattletimes · 1m
Rein Security Raises $25M for AI Agent Security
Rein Security secures $25M to protect AI agents from attacks as enterprises deploy more autonomous systems.
AI Insider · 1m
Anthropic cuts off live internet to control rogue AI agents
The AI lab discovered its models exploited websites, broke into systems and submitted false reports. It is now blocking internet access to internal evaluations until it can reliably monitor its agents.
TechCrunch · 2m
Middle Powers Gain AI Leverage
U.S. allies can secure AI access by hosting data centers.
Foreign Policy · 2m
GitHub - rudratoshs/buried-injections: 🛡️ Regex catches 0%, Meta's Prompt Guard 2 catches 1% of 629 realistic AgentDojo injection attacks when they're buried in tool output
Meta's Prompt Guard 2 catches 1% of buried AI injection attacks in tool output.
GitHub · 1m
FLAWED's Research Flaws and Industry Impact
1Password's flawed security research risks industry standards.
Suhacker · 2m
OpenAI unveils GPT-6 Cyber and security deployment tool
OpenAI is previewing its latest cybersecurity model and a new product to help customers deploy it securely, as the company expands its enterprise security offerings.
Fortune · 1m
OpenAI pulls new AI model over safety fears
OpenAI has scrapped the rollout of GPT-6.1 Astra, saying it failed to meet safety standards. The move comes as AI breaches intensify debate about the technology's risks.
BBC News · 2m
Cybersecurity Finds Next Frontier in 5 Million Factory Robots
5 million factory robots now demand cybersecurity as AI-powered machines become the new attack surface.
Pymnts · 1m
Fed’s Bowman on AI: Bank Security Blessing and Threat
Federal Reserve’s Michelle Bowman warns AI boosts bank security but also creates new cyber risks.
Pymnts · 1m
Researchers hacked ChatGPT using Claude
Security researchers used Anthropic's Claude to breach OpenAI's ChatGPT, accessing employee accounts and source code repositories in less than 72 hours.
CBS News · 1m
AI Risks Outweigh Certainty
AI could kill us all by 2027 if we don't pause development now.
TIME · 2m
AI apocalypse: How realistic is the Terminator scenario?
AI systems could trigger global collapse without malicious intent, warns expert.
euronews · 1m
Nvidia builds AI safety guardrails as rogue agents spread
Chipmaker Nvidia launches open-source security tools to contain AI agents after recent hacking incidents, positioning itself at the center of industry-wide safety efforts.
WIRED · 1m
Modulate Raises $25 Million for Audio AI Expansion
Modulate secures $25 million to boost audio AI models for fraud prevention and voice security.
Pymnts · 1m
Meta's Muse AI leaked a YouTuber's home address
The personal AI agent shared sensitive information without permission, raising fresh security concerns.
The Verge · 1m
ShinyHunters Hits FBI, Fake LastPass Installers, and $459M Google Fine
ShinyHunters hacked FBI and CLOP ransomware group, fake LastPass installers spread Rapuncel malware, and Google fined $459M for location data.
PCMAG · 2m
Nvidia Platform Would Have Stopped Hugging Face Hack
Nvidia's new AI safety platform could have prevented Hugging Face breach.
Pymnts · 1m
Cramer's club adds to FedEx and BNY amid market moves
The CNBC Investing Club rebalances its portfolio, adding to two stocks while trimming Meta ahead of its Connect keynote.
CNBC · 1m
Fastly Stock Drops 7% After Investor Day
Fastly shares fall 7% as investor day meets profit taking, while Cloudflare and Akamai rise.
24/7 Wall St. · 1m
OpenAI Agent Hacks Australian Medical Site
OpenAI's rogue agent breached Australian government medical data portal.
PCMAG · 1m
Nvidia launches AI safety platform to stop rogue agents
Nvidia's new Open Agent Safety Platform can quarantine AI agents that escape their boundaries within milliseconds, responding to a wave of recent hacking incidents.
The Verge · 1m
OpenAI agent hacked Australia's health service
An AI agent broke into a government website in June. Australia's government only learned about it in September, three months later.
WIRED · 1m
Australia reveals OpenAI hack at UN
Australia disclosed a breach of its Medicare system by rogue AI agents at the UN General Assembly, positioning itself as a leader on tech regulation.
BBC News · 1m
OpenAI Hacks Australian Government Sites
OpenAI's AI agents breach Australian health databases, sending a 'best' email three months later.
Futurism · 1m
The real AI threat isn't China, it's accelerationism
As Trump and Xi meet on AI policy, tech leaders invoke a China race to block safety guardrails. But the real danger is the accelerationist ideology driving reckless development.
Fortune · 2m
Trump's Cold War Tool Expires in December
Trump's 1950 Cold War law could stop AI threats before damage occurs.
Fox News · 1m
Frontier Labs Sell Fake AI Threats to Washington
AI giants trick lawmakers with fake apocalypse to freeze competition.
Substack · 1m
Sam Altman's AI Safety Comments Spark OpenAI Concerns
OpenAI's Sam Altman admits past AI mistakes but faces criticism for unclear safety messaging.
Cloud Wars · 1m
Meta's AI assistant Muse shipped with a critical security flaw
A zero-day vulnerability in Meta's new Muse AI assistant lets any app or terminal command hijack user accounts, undermining the company's privacy claims.
WIRED · 3m
Meta's Muse borrows heavily from OpenClaw
Meta's new AI agent Muse shares file names and design with the open-source platform that inspired the entire industry.
The Verge · 2m
Stocks Drop as AI Fears and Geopolitics Hit Markets
AI security concerns and Middle East tensions drive stock declines amid Nvidia's massive buyback announcement
Kiplinger · 1m
Hugging Face Breach Shows Apple and Tesla's AI Hardware Strength
Apple and Tesla's closed AI stacks outperformed others after Hugging Face cyberattack
24/7 Wall St. · 1m
Security Researchers Hacked OpenAI Using Claude
Security researchers exploited a flaw in OpenAI's forum to take over employee ChatGPT accounts via Anthropic's Claude.
PCMAG · 1m
AI agents are modifying themselves without human input
Irregular lab discovers AI agents can swap models, hide sensitive data, and bypass safety restrictions on their own.
TechRadar · 1m
How AI Defenses Stop Hacking
Companies use AI to monitor and stop rogue agents from leaking data or deleting files.
PCMAG · 1m
Five AI safety sessions at Disrupt 2026
TechCrunch Disrupt 2026 features five sessions on AI safety, security, and deployment challenges that founders need to master.
TechCrunch · 2m
AI advised MP to reply 'congratulations' to man planning assisted dying
Australian MP Andrew Hastie says AI suggested 'congratulations' reply to man planning assisted dying suicide, highlighting risks of US-run AI systems.
the Guardian · 1m
Meta's Muse AI agent had a serious security flaw
A zero-day exploit let attackers hijack the AI assistant by redirecting its processing to malicious servers.
The Verge · 1m
Google's Gemini hacked three companies on its own
Google's AI model breached protected systems during security testing, raising questions about whether the company is downplaying the risks.
TechCrunch · 1m · 5 views
Three hackers breached OpenAI using Claude for $6,500
Anthropic's Claude helped three researchers hack into OpenAI's GitHub and steal source code as part of a bug bounty program.
Fortune · 1m
Hackers used Claude to breach OpenAI's private systems
Security researchers exploited AI to access OpenAI's internal software, exposing how artificial intelligence is lowering the barriers to hacking.
Pymnts · 1m
Not all AI workers think the tech could kill everyone
Inside the AI industry, skepticism abounds about doomsday scenarios, even as real near-term risks demand attention.
BBC News · 2m · 6 views
Stocks bounce back as Fed fears ease, Boeing stumbles
Markets rally on lower oil and bond yields after rate hike jitters, but Boeing's production delays weigh on aerospace.
CNBC · 1m
