Amol Kapoor, CEO of Nori Agentic, presents a compelling case for HTML as the universal language for AI agents tasked with creating visual content. In his talk, “HTML is All You Need (for Agents to Make Graphics),” Kapoor challenges the perception that AI agents are solely code-writing entities. He argues that these agents can produce a wide array of visual artifacts, from slides and documents to entire videos, by leveraging the right tools and formats.
HTML: The Key for AI Agents to Create Graphics — from AI Engineer
Visual TL;DR. Pixel-based tools inefficient for AI contrasts with HTML is AI’s language. AI excels at language/structure aligns with HTML is AI’s language. HTML is AI’s language enables AI creates diverse visuals. AI creates diverse visuals leads to Reclaim time. AI creates diverse visuals leads to Enhance creativity. HTML is AI’s language shapes Future of AI visuals.
Pixel-based tools inefficient for AI: human-centric tools like PowerPoint rely on direct manipulationAI excels at language/structure: AI models process information inherently language and structure-basedHTML is AI’s language: universal language for AI agents to create visual contentAI creates diverse visuals: slides, documents, and even entire videosReclaim time: automating visual creation tasksEnhance creativity: focus on higher-level design conceptsFuture of AI visuals: leveraging AI’s natural strengths for graphicsVisual TL;DRQuickExplainDeeper
Visual TL;DR, startuphub.ai
Pixel-based tools inefficient for AI contrasts with HTML is AI’s language. AI excels at language/structure aligns with HTML is AI’s language. HTML is AI’s language enables AI creates diverse visuals
contrasts with
aligns with
enables
Pixel-based tools inefficient for AI
AI excels at language/structure
HTML is AI’s language
AI creates diverse visuals
From startuphub.ai · The publishers behind this format
Visual TL;DR, startuphub.ai
Pixel-based tools inefficient for AI contrasts with HTML is AI’s language. AI excels at language/structure aligns with HTML is AI’s language. HTML is AI’s language enables AI creates diverse visuals
contrasts with
aligns with
enables
Pixel-based toolsinefficient for…
AI excels atlanguage/structure
HTML is AI’slanguage
AI createsdiverse visuals
From startuphub.ai · The publishers behind this format
Visual TL;DR, startuphub.ai
Pixel-based tools inefficient for AI contrasts with HTML is AI’s language. AI excels at language/structure aligns with HTML is AI’s language. HTML is AI’s language enables AI creates diverse visuals
contrasts with
aligns with
enables
Pixel-based tools inefficient for AI
human-centric tools like PowerPoint relyon direct manipulation
AI excels at language/structure
AI models process information inherentlylanguage and structure-based
HTML is AI’s language
universal language for AI agents to createvisual content
AI creates diverse visuals
slides, documents, and even entire videos
From startuphub.ai · The publishers behind this format
Visual TL;DR, startuphub.ai
Pixel-based tools inefficient for AI contrasts with HTML is AI’s language. AI excels at language/structure aligns with HTML is AI’s language. HTML is AI’s language enables AI creates diverse visuals
contrasts with
aligns with
enables
Pixel-based toolsinefficient for…
human-centric toolslike PowerPointrely on direct…
AI excels atlanguage/structure
AI models processinformationinherently language…
HTML is AI’slanguage
universal languagefor AI agents tocreate visual…
AI createsdiverse visuals
slides, documents,and even entirevideos
From startuphub.ai · The publishers behind this format
Visual TL;DR, startuphub.ai
Pixel-based tools inefficient for AI contrasts with HTML is AI’s language. AI excels at language/structure aligns with HTML is AI’s language. HTML is AI’s language enables AI creates diverse visuals. AI creates diverse visuals leads to Reclaim time. AI creates diverse visuals leads to Enhance creativity. HTML is AI’s language shapes Future of AI visuals
contrasts with
aligns with
enables
leads to
leads to
shapes
Pixel-based tools inefficient for AI
human-centric tools like PowerPoint relyon direct manipulation
AI excels at language/structure
AI models process information inherentlylanguage and structure-based
HTML is AI’s language
universal language for AI agents to createvisual content
AI creates diverse visuals
slides, documents, and even entire videos
Reclaim time
automating visual creation tasks
Enhance creativity
focus on higher-level design concepts
Future of AI visuals
leveraging AI’s natural strengths forgraphics
From startuphub.ai · The publishers behind this format
Visual TL;DR, startuphub.ai
Pixel-based tools inefficient for AI contrasts with HTML is AI’s language. AI excels at language/structure aligns with HTML is AI’s language. HTML is AI’s language enables AI creates diverse visuals. AI creates diverse visuals leads to Reclaim time. AI creates diverse visuals leads to Enhance creativity. HTML is AI’s language shapes Future of AI visuals
contrasts with
aligns with
enables
leads to
leads to
shapes
Pixel-based toolsinefficient for…
human-centric toolslike PowerPointrely on direct…
AI excels atlanguage/structure
AI models processinformationinherently language…
HTML is AI’slanguage
universal languagefor AI agents tocreate visual…
AI createsdiverse visuals
slides, documents,and even entirevideos
Reclaim time
automating visualcreation tasks
Enhancecreativity
focus onhigher-level designconcepts
Future of AIvisuals
leveraging AI’snatural strengthsfor graphics
From startuphub.ai · The publishers behind this format
The Limitations of Pixel-Based Creation for AI
Kapoor begins by highlighting the inefficiencies and limitations of current visual creation tools when used by AI. He points out that software like PowerPoint, Google Slides, Figma, and Canva are built with human interaction in mind, relying on direct manipulation through clicks, drags, and resizes. This graphical, pixel-based approach is fundamentally at odds with how AI models process information, which is inherently language- and structure-based.
He illustrates the point by referencing Simon Willison’s test, which asks AI models to generate an SVG of a pelican riding a bicycle. While models can often produce the SVG code, the visual output is frequently flawed, demonstrating a lack of spatial reasoning. Kapoor asserts that this isn’t a failure of the AI models themselves, but rather a mismatch in the medium. Asking an AI to create graphics using pixel manipulation is akin to asking a human to draw complex vector graphics purely by hand, it’s inefficient and prone to error.