xAI preparing updates for Grok, including Grok 3.5 release and new features

xAI is preparing key updates for Grok, including the imminent release of Grok 3.5 and Grok 4 later this year. New features in development span voice mode with camera input, memory recall, Google Drive integration, and image editing tools.

🗞 #grok

Perplexity adds Gemini 2.5 Pro and voice mode to web platform

Perplexity added Gemini 2.5 Pro and Gemini 2.0 Flash to its model selector, introduced voice mode on web, and improved hotel booking with a refined UI. A new Comet browser release is anticipated in the coming weeks.

🗞 #perplexity

Anthropic begins testing voice mode with three voices in Claude App

Anthropic is testing a voice mode in the Claude iOS app with three voice options and English-only support. It also adds file upload features and hints at upcoming integrations with Gmail, Calendar, and Drive, plus UI refinements.

🗞 #claude

OpenAI debuts GPT-4.1 family offering 1M token context window

OpenAI releases GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano via API, offering improved coding performance, instruction adherence, and 1 million token context capacity. These models outperform GPT-4o and are priced lower, with GPT-4.1 nano optimized for speed.

🗞 #chatgpt

Perplexity Sonar model ties with Google Gemini in top AI search ranking

Perplexity’s Sonar-Reasoning-Pro-High model tied for first in LM Arena Search Arena 2025, surpassing Google and OpenAI in key metrics. It leads in depth, source citation, and user preferences, solidifying Perplexity’s leadership in AI search.

🗞 #perplexity

Gemini Advanced subscribers can now generate videos with Veo 2

Google is integrating its Veo 2 video AI model into Gemini Advanced for subscribers, allowing creation of short 720p videos. Veo 2 is also available in Google Labs' Whisk tool, supporting video generation from images via the new Animate feature.

🗞 #gemini

Anthropic adds Research tools and Google Workspace integration to Claude AI

Anthropic has updated Claude with a Research feature for automated, multi-step query analysis and a Google Workspace integration for accessing Gmail, Calendar, and Docs. Both tools support productivity and are available in select plans and regions.

🗞 #claude

OpenAI’s o3 and o4‑mini bring smarter tools and faster reasoning to ChatGPT

OpenAI released o3 and o4‑mini, models that complete complex tasks in under a minute and process visual inputs during reasoning. O3 reduces major errors by 20% over o1, while o4‑mini offers faster performance and high accuracy with tools.

🗞 #chatgpt

Google launches Gemini 2.5 Flash model with hybrid reasoning

Google released Gemini 2.5 Flash, a multimodal developer model with controllable reasoning for cost-efficient, high-volume tasks. It supports up to one million tokens and offers flexible pricing based on reasoning usage.

🗞 #gemini

xAI integrates camera functionality into Grok voice mode for iOS

xAI has begun deploying a vision feature for Grok's voice mode on iOS, allowing camera input for voice-based visual denoscriptions. Full analysis functions are still in development, with availability limited to the latest app version.

🗞 #grok

Microsoft tests new Copilot voices and teases fourth avatar in latest builds

Recent Copilot builds reveal a redesigned Erin avatar, early signs of a new Ellie avatar, and the addition of Alder and Elm voice options. These updates support Microsoft’s approach of gradually shaping Copilot's identity across platforms.

🗞 #microsoftcopilot

Microsoft launches Copilot Merchant Program to integrate retailers into AI shopping

Microsoft launched the Copilot Merchant Program to allow third-party retailers to integrate product catalogs into its AI platform, providing real-time suggestions, price alerts, and in-app purchases, aligning with its broader AI commerce strategy.

🗞 #microsoftcopilot

Gemini’s Scheduled Actions may offer AI task scheduling

Google is testing a Scheduled Actions feature in its Gemini app, allowing users to automate tasks like daily news delivery or image generation. The rollout timeline and whether it's tied to paid tiers remain unclear.

🗞 #gemini

OpenAI prepares Custom GPTs for native GPT-4o image generation

OpenAI is preparing to enable image generation using the GPT‑4o model in Custom GPTs, replacing DALL‑E 3. This change will allow user-built bots to create images directly within conversations using a unified multimodal model.

🗞 #chatgpt

Microsoft Copilot AI Podcast feature enters early testing

Microsoft is adding AI-generated podcasts to Copilot, enabling users to convert topics or documents into noscripted dialogues between synthetic hosts. The feature supports real-time queries and will roll out across platforms later this year.

🗞 #microsoftcopilot

OpenAI and Shopify poised for partnership as ChatGPT adds in-chat shopping

OpenAI is integrating a native shopping flow into ChatGPT, enabling users to complete purchases via Shopify checkout within the chat. This positions ChatGPT as a full-funnel commerce platform with minimal setup for Shopify merchants.

🗞 #chatgpt

ShengShu rolls out Vidu Q1 with full-stack AI video and audio generation tools

ShengShu Technology’s Vidu Q1 lets users generate 5-second 1080p clips with audio from two images and a prompt, delivering render coherence and integrated sound tools that surpass current rivals in speed and accuracy.

🗞 #ai

Google AI Studio experiments with live UI preview for Gemini apps

Google AI Studio is testing a hidden canvas rendering feature responding to developer feedback, allowing live UI previews from Gemini-generated code. A new "Usage" tab is also in development to show API consumption metrics directly.

🗞 #aistudio

New AvatarFX model lets Character.AI users create realistic avatars

Character.AI introduced Avatar FX, a model generating photorealistic video from a single image with synchronized speech, facial, and body movements. It supports long clips, dialogue, and user-directed scenes, now live in labs and coming to the app.

🗞 #characterai

OpenAI expands o3 capabilities with higher limits and task scheduling

User reports suggest OpenAI has raised the ChatGPT o3 model’s usage quota to 50–100 messages per day. Both o3 and o4-mini now support task scheduling, signaling broader integration ahead of a possible “o3 Pro” tier launch.

🗞 #chatgpt

Google readies native image generation in Gemini ahead of possible I/O reveal

Google is preparing to launch native image generation in Gemini, with backend updates and content rights warnings hinting at an imminent release. A public reveal may coincide with Google I/O, though no official timeline has been confirmed.

🗞 #gemini