Latest AI News

In June 2025, Alibaba introduced DeepResearch, a new agent-like mode within their revamped Quark app, marking the company's most significant advancement in AI agent technology. This feature transforms Quark from a cloud storage and search app into a comprehensive AI search tool, leveraging Alibaba's powerful Qwen model series. The development builds on Alibaba's strong position in China's AI landscape, where their Qwen models have become the most widely adopted open-source AI series globally.

Google has released Gemini 2.5 Pro to developers and enterprises, introducing an experimental Deep Think reasoning mode specifically designed for complex math and coding tasks. The new model family features advanced security safeguards that significantly increase protection against indirect prompt injection attacks during tool use, making it Google's most secure model family to date. Project Mariner's computer use capabilities are now available in the Gemini API and Vertex AI, with companies like Automation Anywhere and UiPath already exploring its potential.

Axiom Space is set to launch its fourth mission to the International Space Station on June 10, 2025, carrying astronauts from India, Poland, and Hungary in what CEO Tejpaul Bhatia calls "a little bit of a victory lap." The mission will conduct approximately 60 scientific studies representing 31 countries, making it the most research-intensive Axiom Space mission to date. Among the groundbreaking technologies being tested are AI-powered wearable devices designed to monitor astronauts' sleep quality and readiness for critical tasks.

Apple is preparing to integrate advanced AI technology into its Health app through Project Mulberry, which will feature an AI-powered health coach designed to provide personalized medical guidance based on user data. The virtual health assistant, trained on data from Apple-hired physicians, aims to offer advice similar to what users might receive from real doctors. This initiative represents CEO Tim Cook's vision of healthcare being Apple's greatest contribution to society, with a potential announcement at WWDC 2025 and release expected in 2026.

Apple is set to unveil a major interface overhaul codenamed 'Solarium' at WWDC 2025, representing the most significant redesign across its ecosystem since iOS 7 launched in 2013. The visionOS-inspired interface will feature translucent, glass-like elements across iOS, iPadOS, macOS, watchOS, and tvOS, creating a more cohesive visual experience. While design takes center stage, Apple will also introduce select AI-powered features to enhance the user experience, though competitors continue pushing more aggressively in the AI space.

Apple is set to introduce an innovative AI-powered battery management system at WWDC 2025 on June 9, as part of its iOS 26 update. This Apple Intelligence feature will analyze user behavior patterns to dynamically optimize power consumption, potentially extending battery life significantly. The technology is expected to be particularly crucial for the upcoming ultra-thin iPhone 17 Air, which will feature a smaller battery than standard models.

Apple plans to introduce a centralized gaming app across its ecosystem at WWDC 2025 on June 9, replacing the long-neglected Game Center. The new platform will serve as a comprehensive gaming hub featuring leaderboards, matchmaking, game launching capabilities, and curated editorial content. This strategic move represents Apple's renewed commitment to gaming, potentially creating new revenue streams while enhancing user experience across iPhone, iPad, Mac, and Apple TV.

Meta Platforms is in advanced talks to invest more than $10 billion in Scale AI, a data labeling company crucial for training machine learning models. This potential deal would mark Meta's largest external AI investment and one of the biggest private funding events in history. Scale AI, valued at $14 billion in 2024, has experienced explosive growth with revenue projected to reach $2 billion in 2025, up from $870 million last year.

Scale AI, the data labeling powerhouse behind many leading AI models, is on track to more than double its revenue to $2 billion in 2025. The company, valued at $13.8 billion after its 2024 funding round, has attracted attention from Meta, which is reportedly considering an investment exceeding $10 billion. This growth comes despite recent labor controversies, with the Department of Labor recently dropping its investigation into contractor classification practices.

Meta and Scale AI have unveiled Defense Llama, a specialized version of Meta's Llama large language model designed specifically for military applications. This partnership follows Meta's recent policy shift allowing US defense agencies to use its AI models and comes amid its collaboration with Anduril Industries on an AI-powered military helmet. The move represents a growing trend of mainstream AI companies entering defense partnerships, with OpenAI similarly partnering with Anduril on drone defense systems.

At WWDC 2025 on June 9, Apple announced it will open its 3-billion parameter on-device foundation models to third-party developers. This significant milestone allows developers to integrate Apple's AI capabilities—currently powering features like text summarization and autocorrect—directly into their apps. While less powerful than cloud-based alternatives from competitors, this move expands Apple's AI ecosystem within its privacy-focused framework.

Microsoft Research has developed AI2BMD, a groundbreaking artificial intelligence system that simulates protein dynamics with unprecedented speed and accuracy. This technology enables researchers to explore complex biomolecular problems that were previously intractable, significantly accelerating drug discovery processes. The system has already demonstrated real-world impact by successfully predicting compounds that bind to disease-causing proteins, including the main protease of SARS-CoV-2.

At WWDC 2025 on June 9, Apple introduced 'Liquid Glass,' a significant visual redesign for iOS 26 and all Apple devices, representing the most substantial interface change since 2013. The new design system draws inspiration from Apple Vision Pro's visionOS, featuring translucent, glass-like elements that respond dynamically to content and user input. Additionally, Apple is shifting to a year-based naming convention for its software platforms, with iOS 26 replacing what would have been iOS 19.

OpenAI announced on June 9, 2025, that it has reached $10 billion in annual recurring revenue, nearly doubling from $5.5 billion last year. This milestone comes less than three years after launching ChatGPT, with revenue derived from consumer products, business offerings, and API services, excluding Microsoft licensing deals. Despite impressive growth, the AI giant continues to face significant financial challenges, having lost approximately $5 billion in 2024.

Meta CEO Mark Zuckerberg announced plans to invest up to $65 billion in artificial intelligence infrastructure throughout 2025, marking one of the largest single-year AI investments by any tech company. A significant portion will fund a massive data center in Louisiana that will support Meta's AI initiatives, including the development of its Llama large language models. This investment positions Meta to compete aggressively with rivals like OpenAI and Google in the race to dominate AI technology.

Anthropic has launched Claude Gov, a suite of AI models custom-built for US national security operations. These specialized models are already deployed by high-level security agencies and are designed to better handle classified information while maintaining safety standards. The release marks a significant shift in AI companies' approach to defense contracts, with Anthropic joining competitors like OpenAI, Meta, and Google in pursuing government partnerships.

Apple's AI strategy faced significant scrutiny at its annual Worldwide Developers Conference (WWDC) on June 9, where the company delivered modest AI updates rather than the breakthrough features many had anticipated. After a year of delays and unfulfilled promises with Apple Intelligence, the company opened its AI models to developers but failed to address its lagging position behind competitors like OpenAI, Google, and Meta. Apple services chief Eddy Cue's recent comment that "You may not need an iPhone 10 years from now" underscores the existential threat AI poses to Apple's core business.

Google is now processing a staggering 480 trillion AI tokens monthly across its products and APIs, marking a 50-fold increase from 9.7 trillion tokens just one year ago. The Gemini app has reached 400 million monthly active users, while developer adoption has grown five-fold to 7 million. This extraordinary growth signals a new phase in AI adoption where decades of research are becoming practical reality for users worldwide.

The United Kingdom has established itself as the first nation to criminalize the creation and distribution of AI-generated child sexual abuse material through the landmark Crime and Policing Bill. Introduced in February 2025 and currently progressing through Parliament, the legislation targets those who possess, create, or distribute AI tools designed specifically to generate child sexual abuse imagery. Law enforcement agencies will receive enhanced powers to combat this growing threat, including the ability to search digital devices at UK borders.

Researchers at the University of Hong Kong have developed a groundbreaking drone technology that enables autonomous navigation through complex environments at speeds up to 45 mph, mimicking birds' natural agility. The SUPER system, unveiled on June 7, 2025, uses sophisticated 3D LIDAR sensors and dual-trajectory planning to navigate without pre-mapped routes or GPS. Meanwhile, energy-efficient neuromorphic computing breakthroughs promise to extend AI-enabled drone flight times from minutes to nearly an hour.