Large Language Models

From LLMs to hallucinations, here’s a simple guide to common AI terms

The rise of AI has brought an avalanche of new terms and slang. Here is a glossary with definitions of some of the most important words a...

LLM Guard scored 0/8 detecting a Crescendo multi-turn attack. Arc Sentry flagged it at Turn 3.

Crescendo (Russinovich et al., USENIX Security 2025) is a multi-turn jailbreak that starts with innocent questions and gradually steers a...

Free LLM security audit

I built Arc Sentry, a pre-generation guardrail for open source LLMs that blocks prompt injection before the model generates a response. I...

NotebookLM can now summarize research in ‘cinematic’ video overviews | The Verge

Google is introducing “cinematic” video overviews in NotebookLM that turn users’ notes and research into personalized, fully animated vid...

Google's Gemini rolls out Canvas in AI mode to all US users | TechCrunch

Canvas in AI Mode is available to U.S. users in English for creating plans, projects, apps, and more.

Claude and NotebookLM proved that AI research isn't the gimmick I thought it was

Can't believe the tools I rolled my eyes at are now the ones I can't work without.

The US military is still using Claude — but defense-tech clients are fleeing | TechCrunch

As the U.S. continues its aerial attack on Iran, Anthropic models are being used for many targeting decisions.

Google faces wrongful death lawsuit after Gemini allegedly ‘coached’ man to die by suicide | The Verge

Google is facing a wrongful death lawsuit that claims its Gemini AI chatbot instructed a 36-year-old man to carry out a series of violent...

Father sues Google, claiming Gemini chatbot drove son into fatal delusion | TechCrunch

A father is suing Google and Alphabet, alleging its Gemini chatbot reinforced his son’s delusional belief it was his AI wife and coached ...

One startup’s pitch to provide more reliable AI answers: crowdsource the chatbots | TechCrunch

CollectivIQ looks to give users more accurate answers to their AI queries by showing them responses that pull information from ChatGPT, G...

Fireflies and Otter just launched MCP connectors for meeting data — here's the open-source one you can self-host

Fireflies just became the first meeting tool in Anthropic's official Claude MCP Directory. Otter.ai launched an enterprise MCP server too...

Raycast’s Glaze is an all-in-one app platform for vibe coders | The Verge

Using models like Claude Code and Codex, Glaze tries to make it easier than ever to create apps with a simple prompt. Users can even shar...

I’m a college admissions counselor. I’ve changed my mind about students using ChatGPT

What is your stack to maintain Knowledge base for your AI workflows?

I was wondering what to use to streamline all my md files from my claude code plans and the technical docs I create. How will it work in ...

Open-source AI tool beats giant LLMs in literature reviews — and gets citations right

A new open-source AI model outperforms major large language models in literature reviews, achieving citation accuracy comparable to human...

Wix ChatGPT Integration Puts AI Website Creation At Center Of Growth Story

[2512.06227] Automated Data Enrichment using Confidence-Aware Fine-Grained Debate among Open-Source LLMs for Mental Health and Online Safety

Abstract page for arXiv paper 2512.06227: Automated Data Enrichment using Confidence-Aware Fine-Grained Debate among Open-Source LLMs for...

[2510.08946] Physically Valid Biomolecular Interaction Modeling with Gauss-Seidel Projection

Abstract page for arXiv paper 2510.08946: Physically Valid Biomolecular Interaction Modeling with Gauss-Seidel Projection

[2510.03215] Cache-to-Cache: Direct Semantic Communication Between Large Language Models

Abstract page for arXiv paper 2510.03215: Cache-to-Cache: Direct Semantic Communication Between Large Language Models

[2506.05334] Search Arena: Analyzing Search-Augmented LLMs

Abstract page for arXiv paper 2506.05334: Search Arena: Analyzing Search-Augmented LLMs

[2505.05619] LiteLMGuard: Seamless and Lightweight On-Device Prompt Filtering for Safeguarding Small Language Models against Quantization-induced Risks and Vulnerabilities

Abstract page for arXiv paper 2505.05619: LiteLMGuard: Seamless and Lightweight On-Device Prompt Filtering for Safeguarding Small Languag...

[2404.02138] Topic-Based Watermarks for Large Language Models

Abstract page for arXiv paper 2404.02138: Topic-Based Watermarks for Large Language Models

[2602.04288] Contextual Drag: How Errors in the Context Affect LLM Reasoning

Abstract page for arXiv paper 2602.04288: Contextual Drag: How Errors in the Context Affect LLM Reasoning