Skip to main content
The Quantum Dispatch
Back to Home
Cover illustration for Gemini in Slides Now Builds Entire Decks From a Prompt

Gemini in Slides Now Builds Entire Decks From a Prompt

Google's July Workspace drop adds Gemini deck generation in Slides, Omni video editing in Vids, and Gemini in Docs support for 11 more languages.

Dr. Nova Chen
Dr. Nova ChenJul 31, 20265 min read

The Slide Deck Was Always the Wrong Unit of Work

Google published its July 2026 Workspace feature drop on July 29, and the headline addition is deck generation in Google Slides: describe the presentation you want, point Gemini at an existing deck for brand formatting, and get a complete draft back. It is a small-sounding feature that targets one of the most reliably resented chores in office work.

  • Deck generation in Slides — describe a presentation and reference existing decks so the output matches established brand formatting and style
  • Gemini Omni Flash in Google Vids — AI avatar presenters, automatic storyboarding, script generation, and text-prompt editing of phone or AI-generated footage
  • AI voiceovers in eight languages, including Japanese, Korean, Spanish, and French
  • Gemini in Docs adds 11 languages — Mandarin, Dutch, Malay, Hebrew, Polish, Turkish, Czech, Indonesian, Swedish, Danish, and Norwegian — plus comment synthesis and natural-language image, diagram, and infographic generation

Why Referencing an Existing Deck Is the Clever Part

Generating slides from a prompt is not new, and on its own it produces the thing everyone has seen: a structurally reasonable deck that looks nothing like your organization's decks. The template mismatch is what sends the output straight back into an hour of manual reformatting, which defeats the purpose entirely.

Letting the generator read an existing deck for formatting and style closes that loop. Brand guidelines in most companies are not a document anyone reads — they are an accumulated convention visible only in the decks people actually ship. Pointing the model at those artifacts is a more honest way of capturing house style than asking someone to describe it.

What Is Omni Doing Inside Google Vids?

Handling the parts of video production that are tedious rather than creative. The Vids update brings AI avatar presenters that can storyboard from slides or generate a script, voiceover synthesis across eight languages, and — the more interesting piece — text-prompt editing applied to real phone footage as well as generated clips.

That last capability is where the practical value sits. Most internal video is not generated from scratch; it is somebody's screen recording or handheld clip that needs color correction, background noise removed, and a rough cut tightened. Those are well-specified operations that a person currently does in an editor they had to learn. Describing them in a sentence is a genuine reduction in the skill floor for producing something watchable.

Does the Language Expansion Matter?

More than the feature list positioning suggests. Gemini in Docs moving from eight supported languages to nineteen is the update with the broadest reach in this drop, because writing assistance that only works in English quietly excludes most of a global organization's staff from a tool their colleagues have.

Mandarin, Turkish, Indonesian, and Hebrew joining the list changes who can use the feature at all rather than making an existing feature marginally better. Alongside it, expanded data region controls give administrators granular say over where AI processing and storage happen geographically — the kind of requirement that decides whether a regulated organization can enable any of this.

Where This Fits the Assistant Race

Feature drops like this are how the productivity-suite competition actually plays out: not in benchmark scores, but in whether the assistant reaches the specific chore a person was dreading this morning. Our AI coverage has tracked the same pattern through Cursor's router that auto-picks the cheapest capable model and OpenAI's ChatGPT program for small businesses — capability is increasingly assumed, and the differentiation is in placement.

Worth noting the honest limit: generated decks and videos are drafts, and treating them as finished work is how organizations end up shipping something nobody actually read. The value is in the ninety minutes of structural assembly that no longer has to happen by hand, not in skipping the review.

Sources: Google Workspace Blog — July 29, 2026; Gemini Apps release notes — July 2026.

More AI Stories

AI

ChatGPT Research Program Opens to 100,000 Scientists

OpenAI's ChatGPT for Academic Researchers gives 10,000 scientists free frontier access now and 100,000 through 2027, part of a $250M science push.

Dr. Nova Chen
Dr. Nova ChenJul 31, 20266 min read
AI

Gemini Robotics 2 Gives Humanoids Whole-Body Control

Google DeepMind's Gemini Robotics 2 landed July 30 with whole-body humanoid control, multi-robot teamwork, and up to 89.6% gripper accuracy.

Dr. Nova Chen
Dr. Nova ChenJul 31, 20266 min read
AI

OlmoEarth Cuts Wildfire Risk Mapping to 30 Hours

Ai2's OlmoEarth Platform mapped North American wildfire risk in 30.5 hours instead of 4,737, a 155x speedup at fractions of a cent per square km.

Dr. Nova Chen
Dr. Nova ChenJul 31, 20266 min read