by Kronos Teknoloji

← Back to topics

AI Safety & Cybersecurity

91 stories in this topic.

A digital visualization of an AI neural network undergoing a safety inspection scan to detect hidden capabilities.

New Method Prevents AI Safety Test Sandbagging

May 10, 20261 min readAI Safety & Cybersecurity

A digital blue security shield icon overlaid on a modern government building, representing the US AI vetting agreements with major tech companies.

US Agency Signs AI Vetting Deals with Tech Giants

May 7, 20261 min readAI Safety & Cybersecurity

A modern university research environment with digital screens displaying data visualizations, representing artificial intelligence and cybersecurity innovation.

Professor Victor Chang Wins Top Cybersecurity Award

May 4, 20261 min readAI Safety & Cybersecurity

A digital DNA strand being analyzed by a blue holographic interface in a modern biotechnology laboratory.

Experts Demand DNA Screening to Curb AI Biothreats

May 4, 20261 min readAI Health & Biotechnology

A high-tech digital visualization of complex software code being analyzed by an artificial intelligence system in a glowing blue server room environment.

Anthropic Mythos AI Finds Thousands of Cyber Flaws

April 26, 20261 min readAI Safety & Cybersecurity

An editorial illustration of a digital neural network representing an AI model with a security lock icon being bypassed by a digital figure.

Anthropic Investigates Unauthorized Access to Claude Mythos

April 24, 20261 min readAI Safety & Cybersecurity

A darkened server room with a computer monitor showing a security alert for a system breach.

Anthropic Probes Breach of Unreleased AI Model

April 24, 20261 min readAI Safety & Cybersecurity

A digital illustration representing machine learning vulnerabilities with a glowing network structure showing cracks and red alert symbols.

Generative AI in ML Raises Cyberattack Risks

April 23, 20261 min readAI Safety & Cybersecurity

Digital illustration of a glowing AI neural network contained within a secure glass vault, symbolizing restricted access to high-risk software.

Anthropic Limits Claude Mythos Over Cybersecurity Risks

April 21, 20261 min readAI Safety & Cybersecurity

A digital visualization of computer code with highlighted vulnerabilities overlaid on a map of critical utility infrastructure.

Anthropic Limits Claude Mythos Release Over Cyber Risks

April 16, 20261 min readAI Safety & Cybersecurity

A digital representation of complex computer code glowing on a screen, featuring a translucent butterfly emblem to represent Project Glasswing in a dark server room setting.

Anthropic Restricts Claude Mythos Over Zero-Day Finds

April 13, 20261 min readAI Safety & Cybersecurity

A digital illustration of a glowing blue padlock protecting a complex web of interconnected data points, symbolizing AI cybersecurity and private data protection.

Anthropic Withholds Mythos Model Over Security Risks

April 12, 20261 min readAI Safety & Cybersecurity

A 3D digital illustration of a transparent wing over a glowing circuit network, symbolizing defensive cybersecurity and Anthropic's Project Glasswing.

Anthropic Launches Project Glasswing for Cybersecurity

April 9, 20261 min readAI Safety & Cybersecurity

An artistic representation of a digital brain with an internal biological structure, representing the UCLA study on AI internal embodiment and human safety.

UCLA Study: AI Lacks Human-Like 'Internal Embodiment'

April 4, 20261 min readAI Safety & Cybersecurity

A close-up view of a futuristic quantum computer processor with glowing blue components and complex hardware inside a dark laboratory.

10,000 Qubits Could Break Modern Internet Encryption

April 1, 20261 min readAI Safety & Cybersecurity

A conceptual illustration of a human and robotic hand reaching toward a central glowing node of light, symbolizing cognitive alignment and collaboration between humans and artificial intelligence.

Stevens Researchers Propose AI 'Cognitive Alignment'

March 20, 20261 min readAI Society & Ethics

A smartphone displaying a text-based AI chat session, set against a sterile, out-of-focus medical background.

AI Chatbots May Fuel Delusional Thinking, Study Warns

March 14, 20261 min readAI Safety & Cybersecurity

Conceptual editorial image showing logos of major AI companies grouped together against a backdrop of legal and national security symbols.

Google and OpenAI Scientists Back Anthropic in Lawsuit

March 10, 20261 min readAI Safety & Cybersecurity

Illustration of AI system with warning signs

OpenAI Leak Exposes Toxic Prompt Database and Security Risks

February 18, 20261 min readAI Safety & Cybersecurity