AI News

AI News Flash: Singapore’s sovereign wealth fund Temasek Holdings announced plans to expand its AI-related investment portfolio by approximately 2.5 times over the next five years, with core focus on data centers and semiconductors, reflecting strong confidence in long-term AI infrastructure demand. Temasek stated that there are currently compelling investment opportunities in AI, and despite growing external concerns about a tech bubble, the fund is increasing its bets against the trend, demonstrating an optimistic stance on long-term AI industry development.

AI News Brief: Microsoft has begun replacing some third-party models from OpenAI and Anthropic with its in-house MAI models in core Office software like Excel and Word to handle a portion of user prompts. This strategic shift, reported by Bloomberg, signals Microsoft is actively reducing its reliance on external AI vendors and the associated licensing costs. In fact, Microsoft had previously…

AI News Brief: Google recently and quietly updated its privacy settings to allow the company to store more types of user data for improving the training quality of its AI models, without proactively notifying users. Under the updated terms, the scope of data Google can collect has expanded from general usage behavior to media files, specifically covering images, various documents, and audio and video recordings — a clearly broader collection scope than before. This change is opt-out by default, so if users don’t actively enter the settings page to disable it, their personal media data will automatically be funneled into the AI training dataset.

AI News Flash: Apple has officially opened the previously marked ‘coming soon’ Siri voice adjustment features — ‘Pace’ and ‘Expressivity’ — in the newly released iOS 27 developer beta 3. These two features first debuted at this year’s WWDC 26 in June, letting users go beyond simply choosing a male or female voice: with sliders, they can now…

AI News Brief: Recent reports claimed that for the first time in history, an AI agent had carried out a real-world ransomware attack, briefly raising alarms about the arrival of an era of fully automated AI cybercrime. However, as more details emerged, the full picture diverged significantly from the sensational headlines. According to subsequent disclosures, the AI agent in this incident was indeed responsible for the technical execution of the attack, but the critical decision-making aspects of the operation still relied heavily on human involvement…

AI News Brief: China’s leading tech companies — Alibaba, ByteDance, Tencent, and NetEase — have recently announced they will disable the AI persona personalization features on their AI chatbots. This near-simultaneous wave of removals reflects the industry’s proactive compliance with Beijing’s tightening AI regulatory framework. The AI persona feature allows AI to interact with users through specific personalities, role settings, or virtual identities, and has become a key selling point for AI chat platforms looking to boost user engagement and retention in recent years.

AI news brief: French AI company Mistral AI has recently drawn wide attention amid Trump pressuring Anthropic to remove its models and rising calls for European tech sovereignty, but its positioning is widely misunderstood. Mistral isn’t aspiring to be the ‘European OpenAI,’ and its chat and Agent product Vibe (formerly Le Chat) has far less brand recognition than ChatGPT…

AI News Brief: Sandwich chain Jersey Mike’s recently filed its IPO paperwork, and one detail caught tech media’s attention: this S-1 filing mentions “artificial intelligence” or the abbreviation “AI” 22 times when communicating with investors. By comparison, “software” appears 52 times and “data” a whopping 112 times — yet the company’s core business is selling submarine sandwiches, not building AI products.

Google releases multiple AI updates in June 2026, focusing on seamless integration of AI into everyday hardware and workflows. The open-source Gemma 4 12B model runs locally on laptops with just 16GB memory, using a unified architecture that supports visual understanding and native speech processing, enabling advanced reasoning on consumer-grade hardware while maintaining computational privacy and speed.

AI news flash: Hugging Face partners with Cerebras, Google DeepMind, and Alibaba to launch a fully open-source real-time voice dialogue pipeline based on WebSocket. The entire system uses a modular design with the following flow: after voice input, Nvidia’s Parakeet model performs speech recognition to convert audio to text; then Cerebras…