Can Jarvis Describe an Image That Has No Alt Text?

Yes. Jarvis (getjarvis.eu) can describe an image on your screen even when it has no alt text. Press Cmd+/ on macOS or Ctrl+/ on Windows and ask "describe this image" or "what's in this picture?" Using vision models like frontier models from Anthropic, OpenAI, and Google, Jarvis reads the visible image and produces a clear description, the subject, setting, notable details, any visible text, and even the mood or likely purpose. This is valuable for accessibility when a website or document omits alt text, and for quickly understanding screenshots, diagrams, or photos. Jarvis can read the description aloud via voice mode. It's a screen-aware desktop assistant; free to start, then $16/month, GDPR-aligned, and never trains on your images or screenshots.

Missing alt text leaves images invisible to screen readers and confusing to anyone relying on assistive tech. Jarvis closes that gap on demand: point it at the unlabeled image and it describes what's there in as much or as little detail as you want. Ask for a one-line summary or a thorough description; ask specifically "is there text in this image and what does it say?" It works on photos, infographics, charts, icons, and memes alike, because the same vision model that recognizes objects also reads embedded text and infers context.

Describing images isn't only an accessibility feature. You can use it to caption a screenshot before sharing, to understand a diagram in documentation that's never explained in words, or to identify what an unlabeled icon represents in an unfamiliar app. Designers can even ask Jarvis to draft alt text for images they're publishing, then drop that text into their CMS or a Notion content doc, improving the accessibility of what they ship. Because Jarvis lives in a floating bar over every app, describing an image is always one hotkey away.

Screen-aware AI