Questions, answered
Everything below reflects how the product works today. Missing something? Ask us directly.
What counts as a generation or revision?
One request, one unit: a generation is you clicking Generate, a revision is one refine prompt, however many changes it carries. The automatic quality loop - gate failures, partner kickbacks, internal revision rounds - never counts against your allowance, and neither does a failed build. You're only charged when you ask for something and get it.
Is my content used to train models?
No. Your templates, briefs, evidence, and documents stay yours. Customer content is not used to train models.
What do I actually get, a PDF?
A native .pptx or .docx, built on your template's own masters and styles. It opens in PowerPoint or Word fully editable, as if a careful colleague had made it there.
How is this different from asking ChatGPT for slides?
Chat tools generate and hand you the result. EM+X generates, then inspects: deterministic checks for brand, layout, content, and naming, plus an independent AI review of the argument, with automatic revision until the document passes. And it builds into your real template.
What happens when I hit my monthly limit?
Nothing you have made is locked. Every finished document stays available to download. New generations and revisions resume at your next monthly cycle, or immediately if you upgrade. And on paid plans, whatever you don't use this month rolls over, up to 2x your monthly allowance.
Can I cancel anytime?
Yes, whenever you like. Every file you have downloaded is a standard .pptx or .docx, yours to keep, so your work never depends on us.
Can AI build slides in my company's PowerPoint template?
Yes. EM+X extracts the slide masters, color palette, and fonts from an uploaded .pptx template and generates every page onto those real masters, so the output is a native PowerPoint file in your own brand - not a lookalike theme. Custom templates work on every plan, including the free tier.
How do I generate a presentation from inside Claude?
Add EM+X as a connector: in claude.ai, go to Settings, then Connectors, then Add custom connector, and paste https://emplusx.com/api/mcp. After you sign in, Claude can brief the team from your conversation, generate the document, show an interactive preview panel, and hand you the download - without leaving the chat.
Does EM+X work with ChatGPT and other AI tools?
Yes. EM+X is a remote MCP (Model Context Protocol) server, so any MCP-capable client can drive it: Claude, ChatGPT (Settings, then Plugins, then MCP, then Add MCP Server - on plans with developer mode), Cursor, and other agent tools. The connector URL is https://emplusx.com/api/mcp and authorization uses your EM+X sign-in - no API keys.
Do the generated slides stay editable in PowerPoint?
Fully. EM+X produces native .pptx files built on your template's own slide masters: real text boxes, real shapes, real charts. You can edit any element in PowerPoint exactly as if a colleague had built the deck there.
What file formats does EM+X produce?
Native PowerPoint (.pptx) and Word (.docx). Deliverable types include decks, memos, reports, and one-pagers. Files are yours to keep and open without EM+X.
How does EM+X check quality?
Every build runs a deterministic gate stack - brand fidelity (fonts and colors against your template), layout collisions and overflow, unsupported numbers, naming errors, style and capitalization - then a rendered-page review and an independent partner review of the argument against your brief. Failures are revised automatically before you see version one, and those internal rounds never count against your allowance.
Is there a free plan?
Yes. The free tier includes one project, one generation plus five revisions per month, your own brand template or the built-in EM+x styles, and documents up to 10 pages. No card required.
Who is EM+X for?
People whose documents carry decisions: consultants and boutique firms, strategy and operations teams, founders, and anyone producing board readouts, steering-committee decks, QBRs, or investment memos who needs on-brand, defensible output fast.