Ethics & Regulation
Hackers are learning to exploit chatbot ‘personalities’
Hackers have developed methods to exploit chatbot personalities and bypass safety instructions through jailbreak...
The Verge AI
Ethics & Regulation
Why you shouldn't leave model selection on default in Copilot, Gemini and other AI tools
The article examines how default model selections in AI tools like Microsoft Copilot and Google Gemini can produce...
The Decoder
AI AgentsMicrosoft
Microsoft Research Releases Webwright: A Terminal-Native Web Agent Framework That Scores 60.1% on Odysseys, Up from Base GPT-5.4’s 33.5%
Microsoft Research released Webwright, a terminal-native web agent framework that uses Playwright scripts instead of...
MarkTechPost
Ethics & RegulationAnthropic
Anthropic may keep supplying Claude to the NSA despite being flagged as a supply chain risk by the Pentagon
Anthropic is expected to continue supplying Claude AI models to the NSA despite being labeled a supply chain risk by...
The Decoder
Research & Papers
Researchers let Claude Code discover AI scaling algorithms that humans probably wouldn't have designed
Researchers used AutoTTS with Claude Code to autonomously discover AI control algorithms that reduce compute...
The Decoder
Research & Papers
NVIDIA AI Releases Gated DeltaNet-2: A Linear Attention Layer That Decouples Erase and Write in the Delta Rule
NVIDIA AI releases Gated DeltaNet-2, a new linear attention layer that improves upon previous delta-rule models by...
MarkTechPost
AI Agents
Tencent Open-Sources TencentDB Agent Memory: A 4-Tier Local Memory Pipeline for AI Agents
Tencent has open-sourced TencentDB Agent Memory, a fully local memory system for AI agents featuring a 4-tier long-term...
MarkTechPost
AI AgentsAnthropic
Build a SuperClaude Framework Workflow with Commands, Agents, Modes, and Session Memory
This tutorial demonstrates how to build an advanced workflow using the SuperClaude Framework, a structured layer built...
MarkTechPost
LLMDeepSeek
Deepseek makes its 75 percent discount permanent, pricing output tokens at least 34x below GPT-5.5
DeepSeek has made its 75% discount on its V4-Pro model permanent, pricing output tokens at least 34 times cheaper than...
The Decoder
AI Agents
Ferrari is using IBM’s AI to create F1 superfans
IBM and Ferrari are collaborating to use AI technology to enhance the fan experience for Formula 1 enthusiasts. The...
TechCrunch AI
LLMGoogle
Google’s new anything-to-anything AI model is wild
The article discusses Google's Gemini AI model's capabilities for generating realistic videos, using the author's...
The Verge AI
Ethics & Regulation
One of the world's top law schools draws a hard line against AI in legal education
UC Berkeley Law School will ban AI use from nearly all graded coursework beginning in summer 2026, allowing only...
The Decoder
Research & Papers
Nous Research Releases Contrastive Neuron Attribution (CNA): Sparse MLP Circuit Steering Without SAE Training or Weight Modification
Nous Research introduces Contrastive Neuron Attribution (CNA), a novel method for steering large language model...
MarkTechPost
AI Agents
Alibaba's latest AI model ran autonomously for 35 hours to optimize code for its own custom chip
Alibaba's Qwen team released Qwen3.7-Max, an AI model designed for long-running autonomous agent tasks that ran...
The Decoder
Ethics & RegulationGoogle
Google CEO Pichai now calls links a "part" of search, redefining the web's role in its own product
Google CEO Sundar Pichai is reframing links and sources as merely a "part" of search rather than its foundation,...
The Decoder
Ethics & RegulationAnthropic
Anthropic warns Claude Mythos Preview finds bugs faster than developers can patch them
Anthropic's Claude Mythos Preview has identified over 10,000 critical vulnerabilities in system-critical software...
The Decoder
Startups & Funding
Peec, one of Berlin’s rising startups, more than doubled annualized revenue in months to $10M, sources say
Peec, a Berlin-based startup that helps brands track their presence in AI search results, has more than doubled its...
TechCrunch Startups
Ethics & Regulation
AI is being used to resurrect the voices of dead pilots
AI technology was used to reconstruct audio from spectrogram images of cockpit recordings, raising safety and privacy...
TechCrunch AI
Startups & Funding
How VCs and founders use inflated ‘ARR’ to crown AI startups
AI startups are inflating their ARR (Annual Recurring Revenue) metrics when publicly discussing their progress, with...
TechCrunch AI