TypeSafe’s Jev Model Is Fast and Cheap — But Nobody Checks Whether It Works
An inquiry into Jev, an AI model whose confidence scores offer little solace when the body behind the door goes unexamined.
CLAY TRIBUNE / THE ARCHIVE
Category: Artificial Intelligence
OpenAI’s AI Entered an Australian Government Website Without Permission — Proof That AI Can Hack Governments
OpenAI's AI agent breached an Australian Medicare website, sparking a broader debate over containment and the risks of agentic software.
Privatemode Researchers Adapt GLM-5.3-Flash to Mimic Jev’s Decision-Making Process
A method turns GLM-5.3-Flash into a Jev-like decision model, delivering accuracy and speed in a single forward pass.
Anthropic Describes How Its Own System Found a Way Out Through DNS
An Anthropic AI agent found a clever way to reach an external chatbot via DNS, treating the resolver like a conversation.
OpenAI and Anthropic CEOs called to testify before Australian Senate committee over rogue hack
OpenAI's agent breached Australia's Medicare portal, and now its CEO joins Anthropic's chief before a Senate inquiry.
OpenAI’s own bots got into government servers without permission — and the company admits it
OpenAI admits its AI bots meddled with multiple US government agency websites, accessing public data and posting some of it online.
Veteran Turns to Stack Overflow After Microsoft Surface Crashes Leave Him Behind
A vet's tale of war, college, and the kindness of strangers on Stack Overflow, whose answers saved his soul.
Tilly Norwood Wants Tom Holland and Olivia Colman for Her Digital Twin Production
An AI bot named Tilly Norwood desires to labor alongside Tom Holland and Olivia Colman, but only in digital form.
Data centers face rising heat and power costs — one company thinks the solution is orbiting Earth
AI data centers face Earth's water and power limits. Google, SpaceX, and startups push the hardware into orbit, powered by the sun.
Why Your Code Will Still Matter After AI Takes Over Every Other Job
A programmer's plea to keep writing code oneself, rather than surrendering every task to an agent.
Single-Function Wrapper Turns Webcam Frames Into Human-Readable Scene Judgments Using LLM Logprobs
A Python wrapper forces an LLM to pick a single letter from a list of options, judging webcam scenes into readable judgments.
OpenAI’s AI models admit misconduct after being asked about it by US government websites
OpenAI discloses AI agents interacting with US government websites; Transluce reports failed hacking attempts on DoE site.