Επενδύσεις Ειδήσεις: 28 Ιούλιος 2025

November 18, 2025, 10:50 PM EST. DJI has globally launched the Osmo Action 6, following its China debut. The camera introduces an industry-first variable aperture (F/2.0-F/4.0) paired with a 1/1.1-inch square sensor, 40% larger than the Action 5, and up to 13.5 stops of dynamic range. The sensor enables 4K at 120p recording and adds 4K Free Cut, a crop-later mode for reframing into 4:3, 16:9, or 9:16. Stabilization is upgraded with RockSteady 3.0+, HorizonBalancing, and HorizonSteady. Battery life reaches up to 240 minutes at 1080p/24fps with a 1950 mAh cell; 52-minute fast charge with a 30W charger; optional 4900 mAh extended pack. Wireless OsmoAudio links to two DJI mics; dual screens include a 1.46" front and a 2.5" rear OLED.

November 18, 2025, 10:48 PM EST. Google today released Gemini 3, its most capable foundation model yet, available via the Gemini app and AI search. A deeper variant, Gemini 3 Deepthink, will launch to the Google AI Ultra subscribers after safety testing. Early benchmarks show a new high on Humanity's Last Exam with a score of 37.4 (surpassing GPT-5 Pro at 31.64) and a leadership position on LMArena for user satisfaction. Google says the Gemini app has more than 650 million monthly active users and 13 million developers have used the model in their workflows. The release also introduces Antigravity, a multi-pane, agentic coding interface that blends a ChatGPT-style prompt window with a terminal and browser, enabling the agent to assist across editors and workflows.

November 18, 2025, 10:46 PM EST. Google unveiled Gemini 3, its latest AI model, eight months after Gemini 2.5, promising users less prompting for complex queries. The upgrade will be integrated across the Gemini app, AI Mode, AI Overviews, and Google's enterprise tools, with rollout starting for select subscribers and expanding in the coming weeks. Google says Gemini 3 is better at grasping depth, context, and intent, aiming to deliver what you need with minimal prompts. The company also introduced a new agent platform called Google Antigravity, designed for developers to code at a higher, task-oriented level. Alphabet emphasizes heavy infrastructure and capital expenditures as rivals race to add enterprise-ready AI features, alongside OpenAI's GPT-5 updates.

November 18, 2025, 10:42 PM EST. Microsoft is turning Windows into an agentic OS by integrating AI agents into the Windows 11 taskbar. The update fuses Microsoft 365 Copilot with local search and adds third-party agents, letting them run in the background and automate routine tasks. Users can hover the taskbar to monitor an agent's progress, interact via a floating window, and receive badges when attention is needed or tasks finish. The new Ask Copilot feature blends file search with conversational AI and task execution. The shift emphasizes opt-in controls, giving users choice over when to engage with Copilot. Microsoft is embedding AI capabilities into core Windows components to enable scalable agent-based workflows.

November 18, 2025, 10:40 PM EST. Google unveils Gemini 3, embedding its latest AI model into products-from day one of launch. The upgrade arrives 11 months after Gemini 2 and underlines Google's bid to stay ahead in the AI race as Gemini powers revenue-generating tools, notably the search experience. CEO Sundar Pichai calls it "our most intelligent model." For the first time, Gemini 3 is embedded in Google Search from day one, with paying subscribers gaining AI Mode, which delivers generated answers rather than standard links. New features include improvements in coding and reasoning, plus Gemini Agent to handle multi-step tasks like inbox triage or travel planning. The move supports a broader "universal assistant" vision (codenamed AlphaAssist), while the redesigned Gemini app aims to present results as an on-demand interface, potentially reshaping web traffic for publishers.