
OpenAI announced ChatGPT Images 2.5 with a new Sketch feature allowing users to draw prompts directly in chat, improved image quality with natural lighting and richer textures, better multi-turn editing, and up to 50% reduced latency compared to Images 2.0, now available for ChatGPT, ChatGPT Work, and Codex users across desktop, mobile, and web.
AI-generated summary
OpenAI continues to expand ChatGPT's multimodal capabilities, building on previous image generation features with enhanced user interaction methods.
OpenAI announced ChatGPT Images 2.5 on Tuesday and is adding a new way to tell ChatGPT what you want it to make an image of: by drawing a doodle. With a new feature called Sketch, you can just draw something right inside ChatGPT and then tell ChatGPT how you want it to make an image based on that sketch.
You can activate the Sketch feature by typing @Sketch into your chat box, and a window will pop up where you can make your drawing. I briefly tested it, and it successfully turned my bad doodle of a cat that I drew with my computer mouse into a realistic photo, per my instructions. You can also now leave comments directly on parts of images that you want changed, which I used to make the cat’s eyes green instead of blue.
As for ChatGPT Images 2.5, OpenAI says that it makes images with “more natural lighting and richer textures,” is better at following your editing instructions “across multiple turns,” and that image generation latency has been reduced by “up to 50% compared with Images 2.0.” ChatGPT Images 2.5 is out now for ChatGPT, ChatGPT Work, and Codex users on desktop, mobile, and web, according to OpenAI.
AI outlook — possibilities, not facts
OpenAI will integrate Sketch with voice or video input for multimodal prompting in future updates
Possible · Within months

Grant de Swardt, an independent AI consultant in East Sussex, UK, reported unauthorized token usage on his Claude Max 20x account, which Anthropic traced to a compromised session key used to mint unauthorized OAuth tokens. The company suspended his account, issued a partial refund, and later reinstated it after two weeks, but de Swardt cancelled his subscription due to lack of itemized usage tracking and unresolved security concerns.

Cognition, developer of AI coding assistant Devin, raised $2 billion at a $48 billion valuation just four months after a $26 billion round, with annualized revenue growing from $492 million to $900 million. The round was led by Andreessen Horowitz, Accel, Founders Fund, General Catalyst, and Avenir. Cursor, a competitor, sold to SpaceX for $60 billion in April after talks to raise at $50 billion valuation. Cognition leases costly Nvidia servers and trains its own model to reduce reliance on OpenAI and Anthropic, projecting $4–5 billion revenue by end of 2026. Major customers include Mercedes-Benz, NASA, Goldman Sachs, and Citi.

The Exploration Company (TEC), a European space startup with operations across Germany, France, Luxembourg, Spain and Italy, has raised $450 million in a Series C round described as the largest by a European space company. The funding supports development of its reusable Nyx capsule and Storm rocket engine, with goals to dock with the ISS by November 2028. TEC joins U.S.-based Stoke Space, which is raising a $1 billion Series E, in attracting transatlantic investor interest in reusable spacecraft as alternatives to SpaceX’s growing dominance in low-Earth orbit launches.

Meta introduced Muse, a new personal AI agent for U.S. users that integrates with apps like email, calendars, and payments to perform tasks such as sending emails and booking travel. The launch comes less than two weeks after an $18 billion multistate settlement over social media harms, raising questions about consumer trust given Meta's history of privacy violations and data mishandling.

Apple has scheduled its annual product event for September 9 at 10 AM PDT. The company is expected to unveil its first foldable smartphone, the iPhone Ultra, alongside the iPhone 18 Pro and Pro Max models, marking CEO John Ternus's first major launch.

NYU mathematician Tristan Buckmaster alleges that OpenAI researchers used his private progress to solve the Navier-Stokes existence and smoothness problem, a $1 million Millennium Prize challenge, after learning of his team's specific research approach.