NVIDIA just released the G-Assist Plug-In Builder alongside the experimental launch of Project G-Assist inside the NVIDIA app. Powered by a responsive local small language model running directly on RTX GPUs, the tool lets users create Python and JSON-based plugins to control hardware, trigger game guides, and connect external APIs. It gives PC gamers and creators a genuinely customizable, hands-free on-device copilot.
Microsoft just announced the Microsoft 365 Copilot Wave 2 spring release, introducing an Agent Store alongside two reasoning agents, Researcher and Analyst. Powered by OpenAI deep reasoning models, the agents tackle complex multi-step research and data analysis, while updates also add cross-app search and audio-enabled notebooks. It signals Microsoft's aggressive push into human-agent collaboration.
xAI just rolled out Grok Vision, letting users point their smartphone camera at objects, text, or signs and ask questions in real time. Voice mode also picked up multilingual audio and live search. It catches Grok up with ChatGPT and Gemini, though vision is currently iOS-only.
The Washington Post just inked a strategic deal with OpenAI to bring its reporting directly into ChatGPT search. Users querying current affairs, politics, and tech will now see Post summaries, quotes, and direct article links. With over 500 million weekly active users, OpenAI is aggressively locking down vetted journalism to keep its AI grounded.
xAI just rolled out Grok 3 Mini, bringing built-in reasoning to developers at a fraction of the cost of legacy models. Available now via API alongside the standard Grok 3, the Mini version serves full reasoning traces starting at $0.30 per million input tokens. It puts serious price pressure on rivals while rivaling top-tier reasoning benchmarks.
Rumi researchers just noticed that GPT-o3 and o4-mini were quietly dropping invisible Unicode spaces into longer essays, sparking rumors of a stealth text-watermarking system. OpenAI quickly pushed back, explaining it was simply an odd quirk of large-scale reinforcement learning rather than an intentional tracking feature. The phantom characters have since disappeared in follow-up tests.
Google just showed off its Android XR smartglasses in a live public demo at TED. Powered by Gemini, the hardware sports an in-lens color display and can recall where you left a hotel keycard, translate conversations on the fly, and overlay walking directions. Why it matters: It proves multimodal AI is moving to useful everyday eyewear, though consumers likely have to wait until 2026 to buy a pair.
OpenAI users are turning ChatGPT into a real-life GeoGuessr, using the newly released o3 and o4-mini models to deduce exact locations from random photos. By zooming into tiny visual clues and combining them with web search, the models can identify obscure spots in seconds, sparking serious privacy and doxxing concerns.
Google just released an early preview of Gemini 2.5 Flash, its first hybrid reasoning model that lets developers turn internal thinking on or off. You can set token budgets from 0 to 24,576 to balance quality, latency, and cost, delivering reasoning performance second only to 2.5 Pro.
Grok just rolled out conversation memory to personalize its answers. The assistant can now remember your past interactions when providing advice or recommendations. It basically brings Grok up to speed with rivals offering continuous context across conversations.
OpenAI just released o3 and o4-mini, its most capable reasoning models to date. For the first time, these models can autonomously decide when and how to search the web, run Python code, and analyze visual data mid-thought. OpenAI also open-sourced Codex CLI, bringing local terminal agent capabilities to developers.
Grok just released the first version of Grok Studio, introducing built-in code execution and Google Drive integration. The workspace opens generated documents, code, reports, and browser games in a separate window, letting you collaborate with Grok in real time.
OpenAI just rolled out a new Image Library in ChatGPT, giving Free, Plus, and Pro users a central spot to manage all their AI creations. Located in the left sidebar, it gathers every image made with GPT-4o and supports quick downloads and in-line inpainting tweaks. The catch: deleting a single image currently requires removing the entire chat conversation.
OpenAI is reportedly testing an internal prototype for a social network built around ChatGPT's image generation tools. While CEO Sam Altman has quietly sought feedback from outsiders, it remains unclear if it will launch as a standalone app or inside ChatGPT. The move would unlock crucial real-time user data to help train future frontier models.
Anthropic just gave Claude an agentic Research feature alongside direct integration with Google Docs, Gmail, and Google Calendar. Claude can now cross-reference your internal company files with live web searches to deliver cited answers in minutes. It is currently rolling out in beta for paid tiers.
Anthropic is reportedly getting ready to roll out a new voice mode for Claude as soon as this month. The feature is set to launch with three English voices named Airy, Mellow, and Buttery, directly challenging OpenAI's Advanced Voice Mode in ChatGPT. Anthropic's CPO recently confirmed the team had working internal prototypes.
Perplexity's Sonar-Reasoning-Pro just tied Google's Gemini 2.5 Pro for first place on LM Arena's new Search Arena leaderboard. Perplexity models locked down the top ranks by outperforming OpenAI's search models and citing two to three times more sources, winning 53% of direct head-to-head battles against Gemini.
Netflix is testing a new AI search feature powered by OpenAI tech that suggests titles based on your mood or specific vibe. The opt-in feature is currently live for iOS users in Australia and New Zealand, with a rollout to US subscribers expected in the coming weeks and months.
OpenAI just announced it will officially retire GPT-4 from ChatGPT on April 30, replacing it entirely with GPT-4o. The newer multimodal model beats its predecessor across writing, coding, and STEM tasks, though developers will still keep API access to GPT-4 after the deadline.
Apple is reportedly planning to roll out its delayed, smarter Siri features this fall rather than pushing them to 2026, according to The New York Times. The revamped assistant can handle complex actions like editing and sending photos on command, likely landing right alongside iOS 19. Why it matters: If Apple actually delivers this fall, it gives the company a fighting chance to prove it isn't completely falling behind Google and Meta in consumer AI.
