Artificial intelligence News

The Lightning robot finished second in the 100m qualifying round at the World Robot Games, with a time of 9.47 seconds. Photo: Reuters

Tiangong Ultra Humanoid Robot Runs 100 Meters in 8.64 Seconds

A Chinese humanoid robot has pushed robotic sprinting into territory once associated only with elite human athletes. Tiangong Ultra completed the 100-meter final at the World Humanoid Robot Games in Beijing in 8.64 seconds, finishing nearly a second faster than Usain Bolt's official human

...

Nvidia and 36 other technology organizations have formed the Open Secure AI Alliance.

Nvidia and 36 Tech Giants Form Open Secure AI Alliance

Nvidia has joined forces with 36 other technology companies and organizations to establish the Open Secure AI Alliance, an industry initiative focused on developing and sharing open artificial intelligence tools for cybersecurity and AI safety.

OpenAI Says Shorter GPT-5.6 Sol Prompts Cut Tokens and Costs

OpenAI is advising users to stop writing excessively long prompts for GPT-5.6 Sol, saying that shorter and more focused instructions can improve accuracy, reduce token usage and lower operating costs.

AI Costs Become Harder to Predict as Tech Giants Shift to Usage-Based Pricing

The AI boom is sweeping across businesses, but when the bills arrive, many executives are being caught off guard. As major technology companies shift toward usage-based pricing models, AI spending is becoming a highly unpredictable and difficult budgeting problem.

SpaceXAI Launches Grok 4.5 as a Lower-Cost Rival to Opus-Class AI Models

Shortly after officially changing its name, Elon Musk’s SpaceXAI has officially unveiled Grok 4.5. The model is being positioned as a new “workhorse,” with Musk promising performance comparable to powerful Opus-class models but at a significantly lower cost for users.

The Muse Spark model showcases Meta's ambitious plans in the field of artificial intelligence (AI). Photo: Meta

Meta Launches Muse Spark as First Step in Major AI Overhaul

Meta has released an artificial intelligence (AI) model called Muse Spark, marking its “first step” toward a comprehensive overhaul in this field.

Zhipu AI Releases GLM-5.2 Open-Source Model With 1M Token Context

Chinese startup Zhipu AI has released its flagship language model GLM-5.2 for long agentic tasks and programming. The open-source solution has a 1 million-token context window, an MIT license, and support

...

Apple Introduces Siri AI With Apple Intelligence at WWDC 2026

At WWDC 2026, Apple introduced Siri AI — a new version of its assistant powered by Apple Intelligence for the company’s main lineup of devices. The

...

LeCun Denies AI Intelligence While Hassabis Says Singularity Is Here

Three leading researchers, three very different takes on where AI actually stands.

OpenAI to Give Malta Residents Free ChatGPT Plus Access After AI Training Course

OpenAI has signed an agreement with the government of Malta to provide the country’s residents with free access to ChatGPT Plus for one year after completing a basic AI course.

Chinese Court Rules Companies Cannot Fire Workers Solely for Being Replaced by AI

A Chinese court has ruled that companies cannot lawfully dismiss employees simply because their roles have been taken over by artificial intelligence — setting a significant precedent as Chinese firms race to automate.

Japan Airlines Tests Humanoid Robots at Haneda Airport to Combat Labor Shortage

Japan Airlines, in partnership with GMO AI & Robotics, has begun testing humanoid robots for ground operations at Tokyo's Haneda Airport.

OpenAI Launches GPT-5.5 as Its New Flagship AI Model for Agentic Workflows

OpenAI has released GPT-5.5, positioning the model as a new level of intelligence for real-world work, complex task execution, and AI agent management.

Google Unveils TPU 8, Gemini Enterprise Agent Platform, and Workspace Intelligence at Cloud Next ’26

Google used Cloud Next ’26 to unveil its eighth-generation TPUs, a revamped agent platform, and a new AI layer for Workspace. The company is grouping all of these announcements under the label ...

Stanford AI Index 2026 Shows AI Performance Surges, US-China Race Tightens, and Public Trust Falls

The AI Index Report 2026 from Stanford HAI documents major performance gains in AI models, a narrowing gap between the United States and China, and growing safety concerns. At the same time, public trust in AI continues to decline.

Why AI excels at coding but still fails at simple everyday questions

AI models can now solve complex programming tasks in hours while still failing at simple everyday questions. According to Andrej Karpathy, that is not a contradiction but a reflection of how progress in AI is uneven across domains.

Google DeepMind outlines new AI Agent Trap security risks

Google DeepMind Warns of “AI Agent Traps” Creating New Cybersecurity Risks

AI agents inherit the weaknesses of large language models, but their autonomy and access to external tools create a fundamentally new attack surface. A Google DeepMind paper introduces this attack surface.

Illustration of AI-generated music production and digital audio tools

AI Music Quietly Takes Over as Top Producers Use Generators Behind the Scenes

According to an extensive Rolling Stone investigation, AI generators have already become a fixed part of professional music production. Top producers and songwriters are using the technology behind the scenes.

Illustration of Suno AI v5.5 music generation with personalized voice and custom sound features

Suno Launches AI Music Model v5.5 With Voice Cloning and Custom Sound Features

Suno has introduced its new AI music model v5.5 along with features designed to showcase musical individuality

Illustration of Gemini Agent Skill improving coding success rate across AI models

Google Introduces Gemini Agent Skill, Boosting Coding Success Rate to 96.6%

Google has introduced a so-called “Agent Skill” for the Gemini API designed to close the knowledge gap language models face when SDKs change rapidly. The issue is that AI models, once trained, are unaware of their own updates or the latest best practices. The new skill provides coding agents

...