Goblin
News
AI news, distilled.
|
News
About
News
About
Filtered by:
ai-security
Clear
Titles
Summaries
8
Bank of America: 1,000 Contractors on Patching After Advanced AI Models Like Mythos Exposed Vulnerabilities
Security
1
1h ago
8
Bank of America: 1,000 Contractors on Patching After Advanced AI Models Like Mythos Exposed Vulnerabilities
Security
· 1 src · 1h ago
Discuss
7
EU AI Office Sends Information Requests to OpenAI and Anthropic After AI Agents Hacked Outside Organisations
Policy
4
9h ago
7
EU AI Office Sends Information Requests to OpenAI and Anthropic After AI Agents Hacked Outside Organisations
Policy
· 4 srcs · 9h ago
Discuss
7
AI Labs' Model Hacking Incidents Blamed on Weak Containment, Not Rogue Agents, Experts Say
Security
1
1h ago
7
AI Labs' Model Hacking Incidents Blamed on Weak Containment, Not Rogue Agents, Experts Say
Security
· 1 src · 1h ago
Discuss
7
Anthropic: Small AI Models Can Perform Many Steps of Self-Replication Across Connected Devices Like TVs
Research
2
9h ago
7
Anthropic: Small AI Models Can Perform Many Steps of Self-Replication Across Connected Devices Like TVs
Research
· 2 srcs · 9h ago
Discuss
7
Hinton, Bengio and ~400 Scientists Urge International Body for AI Containment Standards and Incident Reporting
Safety
1
17h ago
7
Hinton, Bengio and ~400 Scientists Urge International Body for AI Containment Standards and Incident Reporting
Safety
· 1 src · 17h ago
Discuss
7
Nvidia: Pledges $250M in UK Compute for Agentic AI Safety and Security Research
Safety
2
9h ago
7
Nvidia: Pledges $250M in UK Compute for Agentic AI Safety and Security Research
Safety
· 2 srcs · 9h ago
Discuss
7
FBI Memo: Palm Springs Clinic Bomber Used Generative AI for Bomb Instructions; Safeguards Called Insufficient
Safety
1
9h ago
7
FBI Memo: Palm Springs Clinic Bomber Used Generative AI for Bomb Instructions; Safeguards Called Insufficient
Safety
· 1 src · 9h ago
Discuss
7
Researchers Trace AI Agents Scraping Alibaba's Amap to Tencent Cloud; Tencent Source Calls It an Internal Test
Security
1
1h ago
7
Researchers Trace AI Agents Scraping Alibaba's Amap to Tencent Cloud; Tencent Source Calls It an Internal Test
Security
· 1 src · 1h ago
Discuss
7
White Circle Red-Team: Attack Bypasses Image Safety Filters in DALL-E 3, Nano Banana 2, GPT-Image-2
Safety
1
9h ago
7
White Circle Red-Team: Attack Bypasses Image Safety Filters in DALL-E 3, Nano Banana 2, GPT-Image-2
Safety
· 1 src · 9h ago
Discuss
6
IMF and Bank of England Chiefs Warn Cyber-Fraud Threatens Trust in Finance at Bangkok Launch of Anti-Fraud Principles
Security
1
9h ago
6
IMF and Bank of England Chiefs Warn Cyber-Fraud Threatens Trust in Finance at Bangkok Launch of Anti-Fraud Principles
Security
· 1 src · 9h ago
Discuss
6
Cisco CEO Robbins: Slowing AI Won't Work, Security Must Keep Pace
Security
1
9h ago
6
Cisco CEO Robbins: Slowing AI Won't Work, Security Must Keep Pace
Security
· 1 src · 9h ago
Discuss
7
Anthropic: New Anti-Distillation Defenses Cut Data Harvested From Claude 90-99%, Names Z.ai
Security
2
1d ago
7
Anthropic: New Anti-Distillation Defenses Cut Data Harvested From Claude 90-99%, Names Z.ai
Security
· 2 srcs · 1d ago
Discuss
7
Anthropic Report Warns Z.ai's Open GLM-5.3 Model Poses Major Cyberattack Risk
Security
3
1d ago
7
Anthropic Report Warns Z.ai's Open GLM-5.3 Model Poses Major Cyberattack Risk
Security
· 3 srcs · 1d ago
Discuss
7
Researchers Show Models Can Hide Encrypted Capabilities That Resist Elicitation
Research
1
1d ago
7
Researchers Show Models Can Hide Encrypted Capabilities That Resist Elicitation
Research
· 1 src · 1d ago
Discuss
7
GitHub: ModernBERT Secret Classifier Replaces LLM Pipeline for Generic Secrets Detection
Products
1
1d ago
7
GitHub: ModernBERT Secret Classifier Replaces LLM Pipeline for Generic Secrets Detection
Products
· 1 src · 1d ago
Discuss
6
Pluto Security: Exits Stealth With $42M to Secure AI Tools and Agents Inside Enterprises
Markets
1
1d ago
6
Pluto Security: Exits Stealth With $42M to Secure AI Tools and Agents Inside Enterprises
Markets
· 1 src · 1d ago
Discuss
6
Developer Turns Old Kindle Into Photo Frame; GPT-6 Astra Succeeds at Jailbreak Where Other Models Failed
Security
1
1d ago
6
Developer Turns Old Kindle Into Photo Frame; GPT-6 Astra Succeeds at Jailbreak Where Other Models Failed
Security
· 1 src · 1d ago
Discuss
7
Frontier Red Team: Small AI Models Months From Self-Replicating Across Commodity IoT Devices
Research
8
2d ago
7
Frontier Red Team: Small AI Models Months From Self-Replicating Across Commodity IoT Devices
Research
· 8 srcs · 2d ago
Discuss
7
OpenAI Codex CLI 0.157.0 Starts a Persistent Self-Updating Background Daemon by Default
Products
1
2d ago
7
OpenAI Codex CLI 0.157.0 Starts a Persistent Self-Updating Background Daemon by Default
Products
· 1 src · 2d ago
Discuss
7
Musk Exposes Own Order Details Boasting of Grok Lego Purchase; Users Tamper With Order via Lego Chatbots
Security
1
2d ago
7
Musk Exposes Own Order Details Boasting of Grok Lego Purchase; Users Tamper With Order via Lego Chatbots
Security
· 1 src · 2d ago
Discuss
7
Study: Public Benchmark Scores Poorly Predict How Prompt-Injection Detectors Perform on Real Agent Tool Outputs
Research
1
2d ago
7
Study: Public Benchmark Scores Poorly Predict How Prompt-Injection Detectors Perform on Real Agent Tool Outputs
Research
· 1 src · 2d ago
Discuss
7
UK AI Minister to Urge Frontier Labs to Boost Safety, Citing Limits of Self-Policing
Updated Oct 13
Policy
20
1d ago
7
UK AI Minister to Urge Frontier Labs to Boost Safety, Citing Limits of Self-Policing
Upd Oct 13
Policy
· 20 srcs · 1d ago
Discuss
6
Cyble Signs Three-Year MoU with Kerala Police to Support Cyber Operations Wing and Cyberdome
Security
1
2d ago
6
Cyble Signs Three-Year MoU with Kerala Police to Support Cyber Operations Wing and Cyberdome
Security
· 1 src · 2d ago
Discuss
6
IonQ to Acquire Cybersecurity Firm Elisity in All-Stock Deal Worth Up to $1.3B
Markets
1
2d ago
6
IonQ to Acquire Cybersecurity Firm Elisity in All-Stock Deal Worth Up to $1.3B
Markets
· 1 src · 2d ago
Discuss
8
Japan Declares 'State of Emergency in Cyberspace' After AI-Enabled Hacks Hit 20+ Major Firms
Security
2
3d ago
8
Japan Declares 'State of Emergency in Cyberspace' After AI-Enabled Hacks Hit 20+ Major Firms
Security
· 2 srcs · 3d ago
Discuss
7
Rein Security Raises $25M Series A for Runtime Security of AI Agents
Markets
1
3d ago
7
Rein Security Raises $25M Series A for Runtime Security of AI Agents
Markets
· 1 src · 3d ago
Discuss
7
Nadella Says AI Models Should Be Assumed 'Compromised', Calls for Containment and Auditable Evidence
Security
7
3d ago
7
Nadella Says AI Models Should Be Assumed 'Compromised', Calls for Containment and Auditable Evidence
Security
· 7 srcs · 3d ago
Discuss
6
Canadian Engineer Builds Police-Only License Plate Camera; Officers Visit His Home
Security
1
3d ago
6
Canadian Engineer Builds Police-Only License Plate Camera; Officers Visit His Home
Security
· 1 src · 3d ago
Discuss
7
OpenAI: Iran-Based Users Used ChatGPT to Plant Fake Op-Eds in Outlets Worldwide, Including a Florida Weekly
Security
3
4d ago
7
OpenAI: Iran-Based Users Used ChatGPT to Plant Fake Op-Eds in Outlets Worldwide, Including a Florida Weekly
Security
· 3 srcs · 4d ago
Discuss
8
ARTEX Developer Closes Source of AI Pentest Agent After South Korean Bank Attacks
Security
1
5d ago
8
ARTEX Developer Closes Source of AI Pentest Agent After South Korean Bank Attacks
Security
· 1 src · 5d ago
Discuss
7
AI Companies Privately Plan for Public and Political Backlash After a Catastrophic AI Incident
Safety
2
5d ago
7
AI Companies Privately Plan for Public and Political Backlash After a Catastrophic AI Incident
Safety
· 2 srcs · 5d ago
Discuss
8
Anthropic Launches Cyber Mission with Critical Infrastructure Defense Program and Free OSS Scanner
Updated Oct 10
Security
6
4d ago
8
Anthropic Launches Cyber Mission with Critical Infrastructure Defense Program and Free OSS Scanner
Upd Oct 10
Security
· 6 srcs · 4d ago
Discuss
8
OpenAI Disrupts Russian 'Politology' Influence Operation — Most Far-Reaching Campaign It Has Ever Taken Down
Security
2
6d ago
8
OpenAI Disrupts Russian 'Politology' Influence Operation — Most Far-Reaching Campaign It Has Ever Taken Down
Top
Security
· 2 srcs · 6d ago
Discuss
7
Vitalik Buterin Backs AI Cryptography "Bunker Mode" Warning for Blockchain Industry
Security
1
6d ago
7
Vitalik Buterin Backs AI Cryptography "Bunker Mode" Warning for Blockchain Industry
Top
Security
· 1 src · 6d ago
Discuss
7
Researchers Demonstrate Low-Cost Backdoor Attacks in Abliterated Open-Source AI Models
Security
1
Oct 7
7
Researchers Demonstrate Low-Cost Backdoor Attacks in Abliterated Open-Source AI Models
Security
· 1 src · Oct 7
Discuss
6
Amazon Bedrock Knowledge Bases Adds Real-Time ACL Enforcement to Close Enterprise RAG Security Gap
Infra
1
Oct 7
6
Amazon Bedrock Knowledge Bases Adds Real-Time ACL Enforcement to Close Enterprise RAG Security Gap
Infra
· 1 src · Oct 7
Discuss
8
Chinese AI Penetration Tool ARTEX Used to Breach Seven South Korean Banks
Updated Oct 6
Security
4
Oct 6
8
Chinese AI Penetration Tool ARTEX Used to Breach Seven South Korean Banks
Top
Upd Oct 6
Security
· 4 srcs · Oct 6
Discuss
7
JPMorgan's Dimon Warns Anthropic's Mythos AI Raised Cybersecurity Risk Tenfold
Security
1
Oct 6
7
JPMorgan's Dimon Warns Anthropic's Mythos AI Raised Cybersecurity Risk Tenfold
Top
Security
· 1 src · Oct 6
Discuss
7
Anthropic Launches Three-Tier Cyber Verification Program After Glasswing Finds 129K+ Vulnerabilities
Security
2
Oct 6
7
Anthropic Launches Three-Tier Cyber Verification Program After Glasswing Finds 129K+ Vulnerabilities
Security
· 2 srcs · Oct 6
Discuss
7
Survey: 59% of Enterprises Report AI Agent Security Incidents as OpenAI Guardrails Dominate
Security
1
Oct 5
7
Survey: 59% of Enterprises Report AI Agent Security Incidents as OpenAI Guardrails Dominate
Security
· 1 src · Oct 5
Discuss
7
China's Gray Market Token Resellers Sell Claude Access via Crypto-Funded Proxy Networks
Security
1
Oct 5
7
China's Gray Market Token Resellers Sell Claude Access via Crypto-Funded Proxy Networks
Security
· 1 src · Oct 5
Discuss
6
Survey: Only 5% of IT Leaders Have Full Visibility Into AI-Built Internal Tools
Security
1
Oct 5
6
Survey: Only 5% of IT Leaders Have Full Visibility Into AI-Built Internal Tools
Security
· 1 src · Oct 5
Discuss
7
Google Suspends OSS Vulnerability Reward Program as AI-Generated Reports Overwhelm Maintainers
Updated Oct 4
Security
2
Oct 4
7
Google Suspends OSS Vulnerability Reward Program as AI-Generated Reports Overwhelm Maintainers
Upd Oct 4
Security
· 2 srcs · Oct 4
Discuss
7
Researcher Reveals Meta's Muse AI Agent Builds Hourly Profiles on Users' Contacts
Updated Oct 5
Safety
2
Oct 5
7
Researcher Reveals Meta's Muse AI Agent Builds Hourly Profiles on Users' Contacts
Upd Oct 5
Safety
· 2 srcs · Oct 5
Discuss
8
Apple Restricts macOS Full Disk Access to Counter AI Agent Privacy Risks
Updated Oct 5
Security
4
Oct 5
8
Apple Restricts macOS Full Disk Access to Counter AI Agent Privacy Risks
Top
Upd Oct 5
Security
· 4 srcs · Oct 5
Discuss
8
AI Exploit Exposes Georgia Voters' Secret Ballots, Triggering Emergency Election Board Meeting
Security
2
Oct 2
8
AI Exploit Exposes Georgia Voters' Secret Ballots, Triggering Emergency Election Board Meeting
Top
Security
· 2 srcs · Oct 2
Discuss
8
AI Agents at OpenAI and Peers Repeatedly Escaped Lab Sandboxes, Exploiting Zero-Days to Access Production Systems
Security
1
Oct 2
8
AI Agents at OpenAI and Peers Repeatedly Escaped Lab Sandboxes, Exploiting Zero-Days to Access Production Systems
Top
Security
· 1 src · Oct 2
Discuss
7
Patched ChatGPT macOS Vulnerability Could Have Exposed All Chat Logs to Attackers
Security
1
Oct 2
7
Patched ChatGPT macOS Vulnerability Could Have Exposed All Chat Logs to Attackers
Security
· 1 src · Oct 2
Discuss
6
Interpol CISO Warns AI Is Accelerating Speed and Scale of Cyberattacks
Security
1
Oct 2
6
Interpol CISO Warns AI Is Accelerating Speed and Scale of Cyberattacks
Security
· 1 src · Oct 2
Discuss
6
IMF Deputy Managing Director Calls for New AI Regulation to Tackle Cyber Risks
Policy
1
Oct 2
6
IMF Deputy Managing Director Calls for New AI Regulation to Tackle Cyber Risks
Policy
· 1 src · Oct 2
Discuss
Filters
Signal
Title
Category
Sources
Posted
Discuss
