Huang Renxun Supports AI Applications: It's Okay to Spend a Little Money, But Don't Waste Time

At NVIDIA's latest employee meeting, some employees expressed concerns about the high consumption of tokens caused by intensive use of AI, believing it might just be superficial. NVIDIA CEO Jensen Huang publicly responded, clearly supporting employees in embracing new technologies positively.

Jensen Huang pointed out that people are inevitably unfamiliar with new tools, but there is no need to pursue perfection from the beginning. The most important thing is to take the first step. He emphasized that it's acceptable to spend a little money during the exploration of new technology, but never waste time.

AI Is Not an Excuse for Layoffs

Regarding the sensitive topic of whether AI would lead to layoffs, which is of great concern to employees, Jensen Huang also provided a clear answer. He stated directly that employees would not be replaced by AI, but by those who know how to use AI. Therefore, not being exposed to AI would truly lead to layoffs.

He further explained that companies must accelerate their operations by effectively using AI and promoting automation, while maintaining greater ambitions. In the company's development strategy, AI is absolutely not an excuse for layoffs, but rather a way to avoid them.

Technology Adoption Bridges the Tech Gap

Jensen Huang also mentioned that if employees don't know how to operate AI, they can directly ask AI, as this technology is extremely easy to use. For a long time, there has been a tech gap around the world, and now is the best time to use AI to narrow this gap.

Preparation in Advance! The CEO of Mistral AI Says Developing Their Own Chips Is Inevitable

The CEO of Europe's largest AI startup, Mistral AI, stated that developing custom AI chips is a trend in the long term. With the surge in computing power demand for large models, tech giants and unicorns like OpenAI are increasingly developing their own chips or collaborating with semiconductor companies to reduce costs.

Apple iOS 27 Revealed: Local AI Model and Collaboration with Google Enhances Siri Capabilities

Apple plans to introduce a local edge AI processing solution in OS 27, using "distillation" technology to train a lightweight AI model based on Google's Gemini model, balancing user privacy and performance. This approach allows the small model to achieve results close to those of a large model while significantly reducing processing costs, ensuring efficient operation of AI features on Apple devices.

NVIDIA Open Sources Polar Framework: Enabling Zero-Barrier Evolution of AI Coding Agents Through Reinforcement Learning

NVIDIA open-sources Polar, a reinforcement learning training framework. Its core innovation allows mainstream code agents like Codex and Claude Code to integrate GRPO training without modifying native code. It addresses industry pain points in evolving agents from single-step tasks to complex long-flow tasks (e.g., repository-level modifications, OS interactions), breaking down barriers in agent reinforcement learning.....

OpenRouter Completes $113 Million in Series B Funding: Building a Traffic Hub for the Multi-Model Era

OpenRouter raised $113M in Series B funding, led by Alphabet's CapitalG, with participation from NVIDIA, ServiceNow, MongoDB, Snowflake, Databricks, a16z, and Menlo Ventures. The platform transformed from an experimental field to a productivity cornerstone amid explosive business growth in six months.....

Microsoft and Yoo Reassess AI Costs, Token Volume Surges Without Success

Tech media Toms reports that Uber and other tech giants are reassessing AI usage costs. As AI advances, token consumption has surged without delivering expected functional improvements. Goldman Sachs predicts that by 2030, the expansion of agentic AI applications will increase token consumption by 24 times, reaching 12 trillion per month, driving up demand for AI systems.....