Large Language Models

ChatGPT Applications OpenAI Chatbots Open Source AI AI Applications Generative AI Training Data Security Risks Ethics

OpenAI’s Jalapeño Chip Claims Major Inference Efficiency and Latency Gains

OpenAI’s InferenceX benchmarks claim the Broadcom‑co‑developed ASIC delivers higher throughput, lower latency, better energy efficiency for interactive AI workloads.

3 ARTICLES

last wk.

Qwen3.8 27B Reportedly Matches GPT‑5.6 and DeepSeek V4 While Running on Consumer GPUs

4 ARTICLES

last mo.

Copilot for Word Can Copy Hidden Prompts Into New Documents

11 ARTICLES

last mo.

Oversight Board Finds Major AI Models Less Likely to Criticize Repressive Governments

8 ARTICLES

last mo.

Apple Holds Talks With PrismML Over Running 27B AI Models on iPhone

8 ARTICLES

last mo.

HalluSquatting Lets Attackers Pre-Register AI Hallucinations to Seed Malicious Packages

7 ARTICLES

2mo ago

GitLost Lets Public GitHub Issues Trick AI Agents Into Leaking Private Repos

15 ARTICLES

2mo ago

JadePuffer: LLM Agent Encrypts 1,342 Nacos Configuration Items

4 ARTICLES

2mo ago

New York Fed Finds Weak Bank Fundamentals Turn Runs Into Failures

4 ARTICLES

2mo ago

Reddit Uses Large Language Models to Fight Stealth AI Marketing

3 ARTICLES

2mo ago

Atwood Says AI Is 'Garbage In, Garbage Out' and Warns About U.S. Media Consolidation

3 ARTICLES

2mo ago

India Faces Strategic AI Dependence Unless It Builds Sovereign LLMs

4 ARTICLES

2mo ago

DeepSeek Open-Sources DSpark and DeepSpec to Speed Model Inference

8 ARTICLES

2mo ago

SearchLeak Turns Microsoft 365 Copilot Into One-Click Data Exfiltration Tool

7 ARTICLES

2mo ago

RAG Shifts From Model Choice to Systems Engineering

4 ARTICLES

3mo ago

Global Fight Over Critical Minerals Sharpens as U.S. and Allies Seek to Break China’s Processing Dominance

Large Language Models

OpenAI’s Jalapeño Chip Claims Major Inference Efficiency and Latency Gains

OpenAI Pushes ChatGPT Work to Bring Agent Power to Office Users

Thomson Reuters Launches Thomson‑1, Its Own Legal and Compliance AI Model

Anthropic Fixes Multi-Model Claude Outage After Spike in Errors

Anthropic-Backed Ode Acquires Casper Studios

Qwen3.8 27B Reportedly Matches GPT‑5.6 and DeepSeek V4 While Running on Consumer GPUs

Copilot for Word Can Copy Hidden Prompts Into New Documents

Oversight Board Finds Major AI Models Less Likely to Criticize Repressive Governments

Apple Holds Talks With PrismML Over Running 27B AI Models on iPhone

HalluSquatting Lets Attackers Pre-Register AI Hallucinations to Seed Malicious Packages

Never miss stories about

Large Language Models

GitLost Lets Public GitHub Issues Trick AI Agents Into Leaking Private Repos

JadePuffer: LLM Agent Encrypts 1,342 Nacos Configuration Items

New York Fed Finds Weak Bank Fundamentals Turn Runs Into Failures

Reddit Uses Large Language Models to Fight Stealth AI Marketing

Atwood Says AI Is 'Garbage In, Garbage Out' and Warns About U.S. Media Consolidation

India Faces Strategic AI Dependence Unless It Builds Sovereign LLMs

DeepSeek Open-Sources DSpark and DeepSpec to Speed Model Inference

SearchLeak Turns Microsoft 365 Copilot Into One-Click Data Exfiltration Tool

RAG Shifts From Model Choice to Systems Engineering

Global Fight Over Critical Minerals Sharpens as U.S. and Allies Seek to Break China’s Processing Dominance

Never miss stories about

Large Language Models