AI Research & Science
Peer-reviewed breakthroughs, university studies, and lab discoveries — explained in plain English. AI Maestro tracks the frontiers of machine learning research, neuroscience meets AI, and the science driving the next wave of intelligent systems.
World models that ignore human beliefs predict the wrong actions, new research shows
A new research paper argues that current world models fail because they ignore human beliefs. Systems like Sora, Genie 3, JEPA, and…
Top stories
Measuring benchmark optimization in speech recognition
20h ago
Meta spends hundreds of millions on Microsoft’s AI services
22h ago
Meet UPDF: A Lightweight Adobe Alternative Built for the Agentic Era
yesterday
A third of web pages published since ChatGPT’s launch show signs of AI authorship, study finds
yesterdayMore ai research & science
Anthropic claims its new Claude Opus 5 delivers near-Fable 5 performance at half the token price
Anthropic says Claude Opus 5 matches the performance of its pricier Fable 5 model…
24 Jul 2026
Team uses AlphaFold AI to redesign gene-editing proteins to make them safer
A couple of decades after the discovery of systems that could selectively target DNA,…
24 Jul 2026
Anthropic launches Opus 5
Anthropic has launched Opus 5, a new version of its flagship model that outperforms…
24 Jul 2026
Anthropic Sonnet 3.5 Sets New Benchmark Standards
Anthropic has released a new AI foundation model today. It is Claude 3.5 Sonnet,…
23 Jul 2026
How AI helps scientists design the next generation of medicines
Designing and developing a new medicine is an expensive, failure-prone scientific challenge. A new…
23 Jul 2026
Meet Gigatoken: A Rust BPE Tokenizer that Encodes Text at 24.53 GB/s, up to 989x Faster than HuggingFace Tokenizers
Gigatoken, a Rust-based tokenizer released by Marcel Rød, a Stanford PhD student, processes text…
23 Jul 2026
Send the arXiv AI-generated slop, get a yearlong vacation from submissions
The arXiv preprint server announced that submitters of AI-generated content containing hallucinations will face…
23 Jul 2026
Are AI labs pelicanmaxxing?
Researcher Dylan Castillo has tested whether major AI labs deliberately train models to generate…
22 Jul 2026
Research-Grade EdgeBench Analysis: AI Agent Benchmarking, Leaderboard Analytics, Scaling Laws, and Evaluation Metrics
The EdgeBench benchmark contains 51 tasks designed to test AI agents across different runtime…
22 Jul 2026