Adobe completes Topaz Labs purchase and keeps the brand
The video and image enhancement company stays a standalone brand, and its CEO Eric Yang joins Adobe's Digital Video and Audio team.
Everything we've published, newest first.
The video and image enhancement company stays a standalone brand, and its CEO Eric Yang joins Adobe's Digital Video and Audio team.
The chat-based editor trims pauses, reorders frames and syncs music on request and arrives in early 2027, alongside a separate Studio tool for testing three cuts of one video.
The promotional rate, live since September 21 with no announced end date, halves the API's per-second price against MiniMax's own published list rate.
The Wall Street Journal reports AI-made attack ads have overtaken conventional smears in Israel's race, with Likud tied to the most AI-flagged campaign posts.
The add-on needs Blender 5.1 or later and runs on Higgsfield's cloud credits, and only some of its seven tools return data you can keep editing.
The livestream paused three hours after its August 31 launch and returned with an upvote system, while fal separately previewed real-time reference video.
The Chicago Laugh Factory footage was unaltered; the false Israel caption spread across X and Instagram and was repeated by AI chatbots, the Jewish Journal reported.
The update, unveiled at Evoto's New York event on September 22, adds a DaVinci Resolve retouching plugin and a $500,000 creator marketplace called Asset Hub.
AFP Fact Check traced a viral clip claiming to show a missile strike at Riyadh's airport to an AI-generated video first posted in March, predating this month's Houthi attacks.
Multiples Alternate Asset Management put in $100 million through preferred shares, and Brahma says it has another $100 million of investor interest still to place.
The Captions maker's free engine drew 935,661 views and 7,885 bookmarks in a day, while its install repository sat at 32 stars.
The open-source project ships no clip generator of its own, and its zero-key path renders a finished narrated video without paying a video API.
The video platform's pay-as-you-go API is billed separately from its monthly subscription plans, and the effective rate depends on how many models you commit to.
The board overturned Meta twice in one ruling and said the company's safeguards against AI deepfakes are consistently and fundamentally inadequate.
The agentic workspace gains two video models alongside Veo 3.1, but Perplexity has published no generation quotas, export formats or licensing terms.
The San Francisco startup says its software layer, measured from living neural activity, speeds up text-to-video inference fivefold, though it will not name the base model.
Vendor threads promise the timeline is over, while the tools doing the actual work are aimed at the boring first pass rather than the cut.
A seven-post thread on a 1,310-follower account shows generated video reflowing to the window, and a straight staircase redrawn as a spiral one.
In OpenAI's customer story, Higgsfield's chief executive credits the model's long-horizon task planning for video ad features that now need one engineer.
CeMAS analysed 38 of the channels behind 1,876 videos published between December 2025 and August 2026, which together drew about 7.55 million views.