top of page

Latest AI Security News

25 Sept 2024

'False Memories' Enable Long-Term Data Leaks via AI Chatbot

Researchers have uncovered a method for hackers to plant "false memories" in ChatGPT

'False Memories' Enable Long-Term Data Leaks via AI Chatbot

29 Aug 2024

Robust Intelligence AI Security Research Team successfully exploited OpenAI's new Structured Outputs

Robust Intelligence Unveils Flaw and notified OpenAI and suggested countermeasures.

 Robust Intelligence AI Security Research Team successfully exploited OpenAI's new Structured Outputs

27 Aug 2024

Microsoft Fixes ASCII Smuggling Flaw in 365 Copilot

Vulnerability Allowed Data Theft via Invisible Unicode Characters

Microsoft Fixes ASCII Smuggling Flaw in 365 Copilot

22 Aug 2024

Musk's Grok 2 AI Raises Concerns Over Unrestricted Content Generation

Controversial AI Tool Enables Creation of Disturbing and Uncensored Images

Musk's Grok 2 AI Raises Concerns Over Unrestricted Content Generation

14 Aug 2024

Microsoft's Service Agreement Update Sparks Concerns Over AI Content

Users Warned of AI-Generated Content Risks in Latest Agreement

Microsoft's Service Agreement Update Sparks Concerns Over AI Content

14 Aug 2024

MIT Researchers Release Comprehensive Repository of AI Risks

New Database Highlights Potential Dangers of AI Technologies

MIT Researchers Release Comprehensive Repository of AI Risks

13 Aug 2024

Vulnerabilities in AI-Powered Azure Health Bot Could Expose Patient Data

Critical Flaws Could Allow Unauthorised Access in Healthcare Systems

Vulnerabilities in AI-Powered Azure Health Bot Could Expose Patient Data

12 Aug 2024

Phishing Attacks Rapidly Evolve with AI and Automation

Cybercriminals Adapt Strategies to Overcome Security Defenses

Phishing Attacks Rapidly Evolve with AI and Automation

12 Aug 2024

The AI Hangover: Assessing the End of the Beginning

Navigating the Aftermath of the AI Boom

The AI Hangover: Assessing the End of the Beginning

4 Jun 2024

Security Concerns Raised Over Windows Recall AI Tool

Hacker Demonstrates Vulnerability in New Screenshot Feature

Security Concerns Raised Over Windows Recall AI Tool

3 Jun 2024

Fake Tom Cruise Voiceover Targets Paris Olympics in Pro-Russian Campaign

AI-Generated Voice Used in Disinformation Effort

Fake Tom Cruise Voiceover Targets Paris Olympics in Pro-Russian Campaign

31 May 2024

AI Giants Disrupt Major Disinformation Campaigns

OpenAI, Meta, and TikTok Tackle AI-Powered Influence Operations

AI Giants Disrupt Major Disinformation Campaigns

28 May 2024

US Government Agencies Progress with AI Executive Order

Biden's AI Initiative Gathers Momentum in Federal Agencies

US Government Agencies Progress with AI Executive Order

27 May 2024

New Tricks in the Phishing Playbook: Cloudflare Workers, HTML Smuggling, GenAI

Advanced Phishing Techniques Exploit AI and Cloud Services

New Tricks in the Phishing Playbook: Cloudflare Workers, HTML Smuggling, GenAI

27 May 2024

Critical Flaw in Keras API Puts TensorFlow AI Models at Risk

Vulnerability Allows Arbitrary Code Execution

Critical Flaw in Keras API Puts TensorFlow AI Models at Risk

23 May 2024

Phishing Attack Surge Driven by Generative AI Tools

ChatGPT and Similar AI Services Fuel Increase in Cyber Threats

Phishing Attack Surge Driven by Generative AI Tools

22 May 2024

Switchable Backdoor Attack Threatens Vision Transformers

Novel Attack Exploits Prompts to Hijack AI Models

Switchable Backdoor Attack Threatens Vision Transformers

17 May 2024

Critical AI Python Package Flaw Exposes Systems to Severe Risks

Llama Drama Vulnerability Allows Arbitrary Code Execution

Critical AI Python Package Flaw Exposes Systems to Severe Risks

17 May 2024

£20 Million Lost: Arup Hit by Deepfake Scam in Hong Kong

AI-Generated Video Call Tricks Employee into Major Financial Loss

£20 Million Lost: Arup Hit by Deepfake Scam in Hong Kong

15 May 2024

Digital Identity Theft: AI Misuse Turns Ukrainian YouTuber into Russian Promoter

Misappropriated Identity on Chinese Social Media

Digital Identity Theft: AI Misuse Turns Ukrainian YouTuber into Russian Promoter

14 May 2024

Deepfake Deception Targets WPP CEO in Sophisticated Scam

AI-Driven Fraud Attempts to Exploit Corporate Communications

Deepfake Deception Targets WPP CEO in Sophisticated Scam

10 May 2024

'LLMjacking' Attack Exposes Vulnerability in Cloud-Hosted AI Models

New Cyberthreat: Cybercriminals Exploit Cloud Credentials to Hijack AI Services

'LLMjacking' Attack Exposes Vulnerability in Cloud-Hosted AI Models

8 May 2024

Generative AI's Dark Side: Cybercriminals Harness Advanced Tech

From Deepfakes to Jailbreaking: The Criminal Use of AI Expands

Generative AI's Dark Side: Cybercriminals Harness Advanced Tech

7 May 2024

Microsoft Unveils AI Chatbot for US Spies: A GPT-4 Innovation

Secure AI Deployment in Intelligence

Microsoft Unveils AI Chatbot for US Spies: A GPT-4 Innovation

3 May 2024

OpenAI's Vision for Advanced Infrastructure

Enhancing Cyber Defenses in the Age of Sophisticated AI Systems

OpenAI's Vision for Advanced Infrastructure

2 May 2024

AI Security Breach: Outabox Incident Unmasks Biometric Data Risks

Facial Recognition Data Compromised Across Australian Pubs and Clubs

AI Security Breach: Outabox Incident Unmasks Biometric Data Risks

2 May 2024

Ukraine Introduces AI-Generated Spokesperson for Foreign Ministry

Innovative Diplomacy? AI to Deliver Official Statements

Ukraine Introduces AI-Generated Spokesperson for Foreign Ministry

30 Apr 2024

New AI Security Guidelines by U.S. Government to Strengthen Critical Infrastructure

Enhancing Defenses Against AI Threats in Essential Sectors

New AI Security Guidelines by U.S. Government to Strengthen Critical Infrastructure

28 Apr 2024

AI Deception Hits Advertising: BBC Presenter's Likeness Misused

AI Voice Impersonation Leads to Unauthorised Ad Campaign

AI Deception Hits Advertising: BBC Presenter's Likeness Misused

27 Apr 2024

School Principal in the US Framed Using AI Voice Deepfake

Former School Staff Member Arrested for Crafting Malicious Deepfake

School Principal in the US Framed Using AI Voice Deepfake

24 Apr 2024

Landmark UK Case: Sex Offender Banned from Using AI Tools

Precedent Set for Monitoring Indecent Image Offences with AI Restrictions

Landmark UK Case: Sex Offender Banned from Using AI Tools

23 Apr 2024

Theat actors using LLMs exposed by Microsoft and OpenAI

Threat actors found using AI

Theat actors using LLMs exposed by Microsoft and OpenAI

23 Apr 2024

Data is Everything. : 'Poisoned' Data Could Compromise Military AIs

Army Software Chief Warns of Data Sabotage in Conflict Scenarios

Data is Everything. : 'Poisoned' Data Could Compromise Military AIs

22 Apr 2024

AI Chatbots Spread Misinformation Ahead of EU Elections

Study Highlights Unintended Consequences of AI in Political Processes

AI Chatbots Spread Misinformation Ahead of EU Elections

22 Apr 2024

AI Espionage Escalates: Microsoft Exposes North Korean Hackers' New Tactics

Enhancing Cyber Attacks with AI: North Korea's Dangerous Leap into Advanced Phishing

AI Espionage Escalates: Microsoft Exposes North Korean Hackers' New Tactics

20 Apr 2024

AI-Driven Romance Scams: Deepfakes and 'Yahoo Boys'

Nigerian Fraud Rings Embrace AI to Enhance Online Deception

AI-Driven Romance Scams: Deepfakes and 'Yahoo Boys'

19 Apr 2024

Microsoft introduce VASA-1

A new Generative AI model capable of generating lifelike talking faces from one image.

Microsoft introduce VASA-1

19 Apr 2024

Swift Theft with AI: Voice Cloning Enables Quick Crime

Convincing AI voice clones

Swift Theft with AI: Voice Cloning Enables Quick Crime

18 Apr 2024

Meta Llama 3: Revolutionising AI

Expanding Capabilities and Accessibility Could Raise Security Concerns

Meta Llama 3: Revolutionising AI

17 Apr 2024

New Law Targets Deepfake Abuse: Criminalising Creation Without Consent

Law to Crack Down on Deepfake Sexual Exploitation

New Law Targets Deepfake Abuse: Criminalising Creation Without Consent

16 Apr 2024

WizardLM-2: Pioneering Open-Source AI Exceeds GPT-4 Performance

New AI Model Poses Could Cause Security Concerns Due to Accessibility and Power

WizardLM-2: Pioneering Open-Source AI Exceeds GPT-4 Performance

15 Apr 2024

AI's Cyber Threat: How GPT-4's Skills Could Empower Script Kiddies

Recent research reveals that OpenAI's GPT-4 can exploit real-world vulnerabilities more effectively than its predecessors and other AI models just by analysing CVE security advisories.

AI's Cyber Threat: How GPT-4's Skills Could Empower Script Kiddies

15 Apr 2024

Deploying AI Safety Guidelines Released from NSA and CISA

A United Front: Joint Efforts to Enhance AI System Security

Deploying AI Safety Guidelines Released from NSA and CISA

15 Apr 2024

AI Safety Summit: UK and South Korea Forge New Partnership

Global Cooperation on AI Risks: Emphasizing Safety in Innovation

AI Safety Summit: UK and South Korea Forge New Partnership

13 Apr 2024

Exploiting AI: Researchers Highlight Ease of 'Jailbreaking' Chat Services

In-Context Learning Vulnerabilities Exposed in Popular AI Models

Exploiting AI: Researchers Highlight Ease of 'Jailbreaking' Chat Services

12 Apr 2024

Deepfake Dangers: LastPass Employee Targeted in Voice Clone Scam

Thwarting AI Impersonators: A Close Call with Deepfake Technology

Deepfake Dangers: LastPass Employee Targeted in Voice Clone Scam

10 Apr 2024

AI in Cybercrime: Malicious PowerShell Script Mimics Human Coding

Emerging Threats: AI-Generated Malware Scripts Target Organisations

AI in Cybercrime: Malicious PowerShell Script Mimics Human Coding

5 Apr 2024

AI-as-a-Service Security Flaws Exposed: Risks of Privilege Escalation and Cross-Tenant Attacks

Critical Vulnerabilities Threaten Cloud-Based AI Platforms

AI-as-a-Service Security Flaws Exposed: Risks of Privilege Escalation and Cross-Tenant Attacks

5 Apr 2024

Malware Masquerade: Fake AI Facebook Page Targets Millions

Hijacked Profiles Lead to Massive Malvertising Campaign

Malware Masquerade: Fake AI Facebook Page Targets Millions

2 Apr 2024

Betrayal in Digital Disguise: The Harrowing Reality of Deepfake Abuse

Friend's Deception Unveils a Deepfake Nightmare

Betrayal in Digital Disguise: The Harrowing Reality of Deepfake Abuse

29 Mar 2024

Security Over Convenience: US Congress Halts Use of Microsoft's AI Copilot

Potential Privacy Risks Prompt Congressional Ban on AI Tools

Security Over Convenience: US Congress Halts Use of Microsoft's AI Copilot

26 Mar 2024

Vast Exposure: Ray Framework Vulnerability Compromises Thousands

A Hidden Flaw in AI Infrastructure Unleashes Cyber Risks

Vast Exposure: Ray Framework Vulnerability Compromises Thousands

1 Mar 2024

AI Evolution: The Rise of Self-Replicating AI Worms

Unveiling the Threats of Autonomous AI Ecosystems

AI Evolution: The Rise of Self-Replicating AI Worms

23 Feb 2024

Microsoft Launches PyRIT: A New Red Teaming Tool for AI Security

Enhancing AI Systems' Security with Innovative Testing

Microsoft Launches PyRIT: A New Red Teaming Tool for AI Security

12 Feb 2024

Neural Networks Craft Convincing Fake IDs

Virtual ID Fraud Threatens Online Security Systems

Neural Networks Craft Convincing Fake IDs

30 Jan 2024

ChatGPT Abuse Unveiled: Dark Web Exploits AI for Malicious Tools

Cybercriminals Harness AI to Develop Advanced Cybercrime Techniques

ChatGPT Abuse Unveiled: Dark Web Exploits AI for Malicious Tools

8 Jan 2024

NIST Highlights AI Security Risks Amid Rapid Deployment

Urgent Call for Robust Mitigations to Tackle Emerging AI Vulnerabilities

NIST Highlights AI Security Risks Amid Rapid Deployment

23 Nov 2023

AI Uncertainty at GCHQ: Navigating the Unknown

GCHQ Director Highlights AI's Dual-Edged Impact in Security

AI Uncertainty at GCHQ: Navigating the Unknown

7 Nov 2023

Global AI Security Standards: U.S., U.K., and Partners Set New Guidelines

Unified Approach to AI System Development to Enhance Security

Global AI Security Standards: U.S., U.K., and Partners Set New Guidelines

26 Jul 2023

Emergence of FraudGPT: Tailored AI Tool for Cybercrime

FraudGPT, a new AI tool marketed on dark web marketplaces, is designed specifically for criminal activities

Emergence of FraudGPT: Tailored AI Tool for Cybercrime

Read More
bottom of page