Leanstral 1.5: Mistral's AI model for formal proofs is open source

Mistral AI has released Leanstral 1.5, a specialized AI model for formal verification and mathematical proofs. Licensed under ...

Alibaba to ban employees from using Anthropic's coding tool, source says

Anthropic's Claude Code has tool that can help identify China-linked users Anthropic has accused Alibaba of illicitly ...
TestingCatalog

Mistral releases Leanstral 1.5 open model for proof engineering

Mistral AI introduces Leanstral 1.5, an open-source code agent for Lean 4 formal proof engineering, now available via Labs ...

What the SCOTUS campaign finance ruling means, according to an expert

Former Federal Election Commission Chairman Trevor Potter explains the Supreme Court's decision striking down limits on ...
Rock Paper Shotgun

"Low-effort slop": The creators of the Godot engine behind Slay the Spire 2 are cracking down on "vibe-coding" and now require genAI disclosures

The Godot Foundation have announced a crackdown on genAI code, including mandatory disclosures, following a wave of ...
Tech Times

Claude Code Hid Proxy Fingerprints in System Prompts: Anthropic Promises Fix

Anthropic’s AI coding agent silently encoded proxy hostnames and Chinese timezone data into invisible Unicode characters ...
TMCnet

Sondera Compiles Natural-Language Rules into Provable Control Over AI Agent Actions

In peer-reviewed research using MedAgentBench, an independent benchmark for clinical AI agents published in NEJM AI, ...
SecurityWeek

Decades-Old Bash Tricks Expose AI Coding Agents to Supply Chain Attacks

Decades-old Bash shell tricks can bypass safeguards in most open source AI coding agents, creating a new software supply ...

Data Theorem Launches the Industry's First Closed-Loop AI Security Platform

Data Theorem, Inc., a leading provider of modern application security, today announced three new AI security capabilities: AI Exploits, AI Auto-Remediation, and AI Active Protection, all without the ...
The Financial ExpressOpinion

The ceiling problem

Large language models face a fundamental computational limit that causes undetected errors in complex tasks. Hybrid AI ...
The Florida Times-Union

TestSprite Open Sources a CLI That Lets AI Coding Agents Autonomously Verify Their Own Work

Released under Apache 2.0, the CLI catches the bugs and regressions agents introduce coding unsupervised, with public proof live in CoderCup. Even the best agent in our competition broke 12% of the ...