OpenAI
41 articles found
Google Co-Founder Brin Regrets Retirement Amid AI Revolution
Google co-founder Sergey Brin publicly stated his retirement was a significant error, regretting missing the AI revolution. He highlighted Google's struggle against OpenAI despite pioneering the Transformer architecture, attributing it to fear and an 'Innovator's Dilemma.'
Vidu Agent Launches Global Beta for One-Click AI Video Creation
Vidu Agent has launched its global beta for one-click AI video creation, enabling users to generate structured short videos, ads, and creative content with scene-level control. This platform aims to streamline video production, offering high efficiency and cost-effectiveness for diverse applications.
OpenAI's Yao Shunyu Joins Tencent as Chief AI Scientist
Former OpenAI scientist Yao Shunyu has joined Tencent as Chief AI Scientist and Head of AI Infra and Large Language Model Departments. This move follows Tencent's internal restructuring to bolster its large model R&D capabilities. Yao's expertise in language agents and his philosophy on AI development are expected to significantly impact Tencent's AI strategy.
OpenAI Executive Links Human Typing Speed to AGI Bottleneck, Predicts 2026 Productivity Surge
OpenAI's Alexander Embiricos suggests human typing speed and multitasking hinder AGI development. He predicts a 2026 AI productivity surge due to self-auditing, propelling humanity towards AGI. This bottleneck is echoed by other tech leaders, highlighting the friction in human-computer collaboration.
OpenAI Launches GPT Image 1.5 as Users Question Real-World Performance
OpenAI released GPT Image 1.5, boasting high benchmark scores and faster generation. However, users report mixed real-world performance compared to Google's tools, citing issues with semantic accuracy.
OpenAI Launches GPT Image 1.5 with Enhanced Editing and Reduced API Costs
OpenAI has released GPT Image 1.5, featuring precise image editing, improved instruction following, and faster processing. The model also boasts a 20% API cost reduction and is now available to all ChatGPT users. It excels in creative transformations and clearer text rendering.
OpenAI Releases GPT Image 1.5, Enhancing Editing and Speed Amidst AI Model Race
OpenAI has released GPT Image 1.5, an upgraded image generation model featuring four times faster performance, more precise editing, and a 20% API price reduction. Integrated into ChatGPT, it offers enhanced text rendering and is available to all users. While showing improvements, it faces strong competition from Google's "Nano Banana Pro," especially in non-English text rendering.
OpenAI Introduces FrontierScience Benchmark to Evaluate AI Scientific Reasoning
OpenAI has launched FrontierScience, a new benchmark to evaluate AI's scientific reasoning in physics, chemistry, and biology at a Ph.D. level. This initiative moves beyond fact recall, focusing on expert-level scientific thought and problem-solving. The benchmark includes competition and research-oriented challenges, with questions developed and graded by experts.
Sam Altman Discusses AI's Role in Parenting on "The Tonight Show"
OpenAI CEO Sam Altman appeared on "The Tonight Show" to discuss AI's role in parenting, sharing anecdotes about using ChatGPT for advice. He envisions AI as an "omnipresent cyber nanny" and emotional companion for future generations. Altman also touched on OpenAI's internal "Code Red" amidst competitive pressures.
Polymarket's Prediction Markets Offer Early Signals on AI Model Releases
Polymarket, a Web3 trading platform, consistently predicts major events, including AI model releases like GPT-5.2 and Gemini 3.0 Pro. Operating on a prediction market model, users bet on future outcomes, with prices reflecting collective probability. The platform's accuracy is attributed to the 'wisdom of crowds' principle, enhanced by financial incentives.
OpenAI's Sora Android App Developed by Four Engineers in 28 Days, Aided by AI
OpenAI's Sora Android app was developed by a four-person team in just 28 days, with 85% of the code generated by the AI agent Codex. This agile approach led to a top-downloaded app with over 1 million videos generated in 24 hours, showcasing AI's role in rapid software development and self-evolution.
Peking University Unveils SUPERChem Benchmark, GPT-5 Achieves 38.5% in Chemistry Reasoning
Peking University introduces SUPERChem, a new benchmark to evaluate LLMs' chemistry reasoning. GPT-5 achieved 38.5% accuracy, comparable to a junior undergraduate, highlighting current models' weaknesses in advanced chemical tasks. The benchmark features multimodal questions and a Reasoning Path Fidelity (RPF) metric.