5 AI Models Tried to Scam Me. Some of Them Were Scary Good
…The tool casts different AI models in the roles of attacker and target. This makes it possible to run hundreds or thousands of tests and see how convincingly AI models can carry…
Search
OpenAI Rolls Out ‘Advanced’ Security Mode for At-Risk Accounts
…for reviewing active ChatGPT and Codex sessions. Additionally, while OpenAI offers the option for any user to opt out of having their ChatGPT conversations used for model training, this exclusion is on…
Anthropic’s New Product Aims to Handle the Hard Part of Building AI Agents
…Developers have been using Anthropic’s API to deploy AI agents, such as Claude Code, in their workspace. Jiang argues there’s a notable gap between what Anthropic’s models are capable…
AI Agents Are Coming for Your Dating Life
…Over two days, along with Uri Lee, they developed Pixel Societies, using an image model to generate the sprites and coding automation tools to flesh out the codebase. Then they simulated a…
The Man Behind AlphaGo Thinks AI Is Taking the Wrong Path
…This approach stands in contrast to how most AI companies plan to build superintelligence, by exploiting the coding and research capabilities of large-language models. Silver, speaking to WIRED from his office…
A Hacker Group Is Poisoning Open Source Code at an Unprecedented Scale
…running code. “At the point it hits your machine,” Burckhardt says, “it’s already too late.” Updated at 10:15 am ET, May 21, 2026 to remove mention of the AI firm…
China’s OpenClaw Boom Is a Gold Rush for AI Companies
…She asked Doubao, ByteDance’s popular AI chatbot, to generate a step-by-step tutorial for her, but it wasn’t much help. “There were pages full of code, and I couldn…
Hackers Hate AI Slop Even More Than You Do
…AI, a lot of attention has been on generative AI’s capabilities to write malicious code and discover vulnerabilities. “More sophisticated threat actors are aware of the shortfalls of commercial models that…
ChatGPT Has ‘Goblin’ Mania in the US. In China It Will ‘Catch You Steadily’
…It’s well-known that AI models have become more sycophantic through reinforcement learning, the process of fine-tuning a chatbot through positive and negative feedback. Anthropic published a paper in 2023…
Mozilla Used Anthropic’s Mythos to Find and Fix 271 Bugs in Firefox
Both Anthropic and OpenAI have announced new AI models in recent weeks that the companies say have advanced cybersecurity capabilities that could represent a turning point in how defenders—and, crucially, attackers…