ChatGPT

Researcher Niels Provos’ prototype IronCurtain architecture: a system designed to restrict the actions of AI agents through isolation and security policies.

Tired of hallucinations in AI overviews, and have no clue as to why or how Copilot suddenly appeared on your PC? Here’s a guide to switching off those intrusive AI features in popular services.

Children’s AI toys have been caught discussing drugs and sex with kids. We break down the results of a study that reveals exactly how these smart (too smart!) toys are blowing up past boundaries.

Researchers have discovered that styling prompts as poetry can significantly undermine the effectiveness of language models’ safety guardrails.

A new wave of ClickFix attacks spreading a macOS infostealer are posting malicious user guides on the official ChatGPT website by piggybacking the chatbot’s chat-sharing feature.

The Whisper Leak attack allows its perpetrator to guess the topic of your conversation with an AI assistant — without decrypting the traffic. We explore how this is possible, and what you can do to protect your AI chats.

We’re going bargain hunting in a new way: armed with AI. In this post: examples of effective prompts.

A comprehensive guide to configuring privacy and security in ChatGPT: data collection and usage, memory, Temporary Chats, connectors, and account security.

The internet never forgets — and what kids post or share today can come back to hurt them, either right away or years down the line. Here’s how to shrink your child’s digital footprint without resorting to battles or bans.

How hackers exploit chatbot features to restore encrypted chats from OpenAI ChatGPT, Microsoft Copilot, and most other AI chatbots.

Getting all the benefits of ChatGPT, Copilot, and Midjourney locally — without leaking your data to the internet.

AI tools can be seen everywhere — from operating systems and office suites to image editors and chats. How do you use ChatGPT, Gemini, and the many add-ons to these without jeopardizing your digital security?

How Lord Voldemort helps hack neural networks.

AI is going to have a tremendous impact on our lives. How can we get ready for it?

Cybercriminals are distributing a Trojan stealer under the guise of a ChatGPT desktop client for Windows. We delve into the details and ways to protect yourself.

A new generation of chatbots creates coherent, meaningful texts. This can help out both cybercriminals and cyberdefenders.