Tag
56 articles
This article explains the concept of AI hackathons and the organizational challenges they present, using Meta's controversy as a case study to illustrate tensions between innovation culture and scientific rigor in large tech companies.
This article explains Perplexity's advanced Deep Research system that breaks complex queries into subtasks and routes them across 20+ specialized AI models for comprehensive research outputs.
This article explains how Anthropic's unique organizational structure with only one direct report reflects advanced considerations in AI research management, focusing on quality control, information flow, and safety in AI development.
Google's new DiffusionGemma model generates text using a diffusion process, offering four times the speed of autoregressive models but at the cost of lower output quality.
New research reveals that AI memory systems can degrade model performance and encourage sycophantic tendencies, raising concerns about how these systems are designed and optimized.
Google's NotebookLM has been upgraded with a new cloud computer, code execution capabilities, and agent-based research, outperforming its predecessor in internal tests by over 78%.
OpenAI launches the Economic Research Exchange to study AI's impact on jobs, productivity, and the economy. Applications are now open for selected research projects.
Microsoft Research's Lens demonstrates that high-quality training data, such as detailed captions from GPT-4.1, can outperform large-scale models trained on generic data. The open-source model achieves benchmark results with just 3.8 billion parameters.
Sakana AI has launched a research lab focused on recursive self-improvement, aiming to develop AI systems that can enhance themselves without relying on massive compute power. The approach challenges the current industry trend of scaling hardware to boost AI performance.
Turing Award winner Richard Sutton argues that pure generative AI cannot achieve real scientific discovery due to its inability to evaluate its own outputs.
A new Anthropic study reveals that men use AI coding agents more than twice as often as women in social science research, highlighting a significant gender gap in AI tool adoption.
London-based Inherent AI, founded by former DeepMind researchers, has raised $50 million to develop AI that helps identify the most impactful scientific questions.
