NewsAnthropic - Mythos cybersecurity model
Mythos | Double-edged sword
Anthropic's Mythos AI model highlights cybersecurity risks by identifying vulnerabilities and generating exploits, raising alarms in critical sectors. thehindu.com
🧠 Large Language Models
News about foundation models, LLMs, multimodal models, benchmarks, and model releases from OpenAI, Anthropic, Google, Meta, Mistral, and others.
463 articles
ResearchLLM applications - healthcare
Evaluating Large Language Models for Automated Data Cleaning and Feature Engineering in Clinical Datasets
Background: Electronic health records (EHRs) are increasingly used for clinical research and machine learning, yet they are plagued by missing values,... cureus.com
OpinionAnthropic - Claude sentience/feelings
The Idea That Claude Has Feelings Is Great for Anthropic
Richard Dawkins is one of the modern world's great skeptics. His 2006 book The God Delusion tore through arguments for the existence of a higher power and... bloomberg.com
ResearchLLM training efficiency - DELIFT
NCSA Resources Enable Development of Data-Efficient LLM Training Method ‘DELIFT’
March 6, 2026 — You may have read about how artificial intelligence applications like ChatGPT were trained by ingesting as much freely available data on the... hpcwire.com
OpinionClaude security capabilities
The vulnerability flood is here. Patching won't save you. | perspective | SC Media
COMMENTARY: Something fundamental shifted on April 7. Anthropic announced Claude Mythos Preview — and disclosed that over 99% of the vulnerabilities the... scworld.com
ResearchLLM inference efficiency
Sakana AI Research Targets More Efficient Large Language Model Inference
According to a recent LinkedIn post from Sakana AI, the company is highlighting research on making large language models faster and more resource efficient... tipranks.com
NewsLLM inference hardware
Optical AI Servers Speed Large Language Model Inference
Lumai's Iris Nova server uses optical computing to run AI models efficiently with low energy consumption. designnews.com
NewsLLM pricing comparison
LLM Pricing: Top 15+ Providers Compared
LLM pricing now splits into two purchases: a per-token API budget for code that calls models programmatically, and an $8–$200 monthly subscription for... aimultiple.com
Launchopen reasoning model release
Meet ZAYA1-8B, a super efficient open reasoning model trained on AMD Instinct MI300 GPUs
The real headline is what ZAYA1-8B was trained on: a full stack of AMD Instinct MI300 graphics processing units (GPUs), the rival to Nvidia GPUs. venturebeat.com
Researchquantum computing + LLM
Multiverse Computing Cuts LLM Perplexity 1.4% On 156-Qubit Processor
Researchers at Multiverse Computing have demonstrated a 1.4 percent improvement in perplexity, a measure of how well a language model predicts a sample,... quantumzeitgeist.com
Researchinterpretability/mechanistic AI
Natural Language Autoencoders: Turning Claude’s thoughts into text
When you talk to an AI model like Claude, you talk to it in words. Internally, Claude processes those words as long lists of numbers, before again producing... anthropic.com
ResearchLLM benchmarks/medical reasoning
New AI model beats doctors at clinical reasoning, diagnosis
A large language model (LLM) matched or exceeded hundreds of expert physicians in diagnostic and management reasoning tasks across six experiments,... msn.com
Researchlong context models/theory
The Inescapable Long Sequence Model Trade-off
A new theoretical framework reveals an inescapable trade-off between efficiency, compactness, and recall in long sequence models. startuphub.ai
Newson-device LLM/Google Chrome
Chrome silently installs a 4 GB local LLM on your computer
You did remember to opt out of AI, didn't you? theregister.com
Newson-device LLM/Google Gemini privacy
Google Chrome Is Downloading a 4GB AI Model Onto Your Device Without Consent, Researcher Warns
Even if you aren't using Google Gemini, it might be using your device. Security researcher Alexander Hanff, also known as “That Privacy Guy,” recently... gizmodo.com
Launchlong-context LLM architecture
Subquadratic: $29 Million Seed Raised For Long-Context AI Architecture
Subquadratic, an AI research company developing long-context large language model architectures, announced the launch of SubQ 1M-Preview, which it describes... pulse2.com
Newssovereign LLM development
B80-million Thai-language AI project launched
The government is preparing to launch a new Thai large language model (LLM) project intended to secure artificial intelligence (AI) sovereignty and drive... bangkokpost.com
Newsnovel LLM architecture claim
Miami startup Subquadratic claims 1,000x AI efficiency gain with SubQ model; researchers demand independent proof.
A little-known Miami-based startup called Subquadratic emerged from stealth on Tuesday with a sweeping claim: that it has built the first large language... venturebeat.com
ResearchLLM inference acceleration on TPUs
Supercharging LLM inference on Google TPUs: Achieving 3X speedups with diffusion-style speculative decoding
The current landscape of Large Language Model (LLM) acceleration is dominated by autoregressive speculative decoding, where a lightweight drafter predicts... developers.googleblog.com
ResearchLLM benchmarking/clinical reasoning
AI outperforms docs on clinical reasoning, but not ready for solo work
New research shows that a large language model outperformed physicians in various clinical reasoning tasks; however, the study's authors cautioned that the... techtarget.com