News

Opera Mini launches AI-powered update to compete with Google and Microsoft...

AI Observer
Apple

Noise-cancelling headphones may be great for the office, but they also...

AI Observer
Meta

WhatsApp tests new widget to make Meta AI more accessible

AI Observer
News

Week in Review: OpenAI may charge $20K per month for an...

AI Observer
News

Copilot could soon get more Microsoft AI, but less ChatGPT

AI Observer
News

Survey finds that ChatGPT is the most popular AI in offices...

AI Observer
News

Exclusive: Startup combines AI with physics to discover new green materials...

AI Observer
AI Regulation & Ethics

Google removes all mentions of a ‘diversity and equity’ from the...

AI Observer
News

AI and human emotions are the building blocks for effective creative...

AI Observer
AI Hardware

Axiom and Red Hat to launch edge computing into space

AI Observer
AI Hardware

McDonald’s invests in AI to boost order accuracy and streamline operations...

AI Observer

Featured

News

OpenAI’s Deep Research is more accurate than you in fact-finding, but...

AI Observer
News

OpenAI releases new simulated reason models with full access to tools

AI Observer
News

xAI adds a memory feature to Grok

AI Observer
AI Hardware

Congress wants to know if Nvidia superchips slipped through Singapore to...

AI Observer
AI Observer

OpenAI’s Deep Research is more accurate than you in fact-finding, but...

Wei and team don't directly offer any hypothesis about why Deep Research fails almost half the time, but the implicit answer is in the scaling of its ability with more compute. As they run more parallel tasks, and ask the model to evaluate multiple answers, the accuracy scales...