🎨 Creative

AI for Creative Tasks

Creating images, enhancing photos, generating videos, writing texts – AI as a creative toolbox.

🤖 AI CompendiumAI for Creative Tasks

The most exciting AI applications for home users are in the creative field. What was science fiction two years ago now costs a few euros a month – or runs free on your PC. Here we show you what you can really do with it.

🖼️

Create AI Images

From a short text to a photorealistic image – that's the world of generative image AI. You describe what you want to see, and the AI creates it in seconds. Useful for hobbies, marketing, concept art, gifts, social media.

The Most Important Tools in 2026

Midjourney

The visual king. Best image quality, often unrivaled aesthetically. Runs on the web and on Discord. Subscription-based. Strong for artistic styles.

DALL·E (in ChatGPT)

OpenAI's image model, integrated into ChatGPT. Very good at understanding concepts and text in images. Practical because it can be controlled via language.

Stable Diffusion

Open source – runs locally on your PC (with a good graphics card). Maximum control, free, no data leaves your device. For professionals and tinkerers.

Adobe Firefly

Adobe's image AI, integrated into Photoshop & Co. Trained only on licensed images – important for commercial use.

How to Formulate Good Prompts

  • Specific instead of vague: Not "a dog", but "a golden Labrador, sitting on a wooden porch, golden evening light"
  • Specify style: "in the style of a watercolor", "photorealistic", "cinematic look", "Pixar style"
  • Describe image composition: "bird's-eye view", "close-up", "mountains in the background"
  • Negative prompts (what you don't want): "no people", "not distorted"

💡 Legal Notice: You automatically have the right to commercial use of AI images – but beware: For tools trained on copyrighted works (Midjourney, Stable Diffusion), there are legal gray areas. Adobe Firefly is on the safe side here because it only uses licensed training images.

📸

Enhance Old Photos with AI

Probably the most useful AI application for home users: restoring old, blurry, or faded photos. What required expensive restorers just a few years ago, AI can now do with a click.

What AI Can Do with Photos

  • Remove blur — AI reconstructs probable sharp details from blurriness
  • Increase resolution — upscale small images without pixelation (Super Resolution)
  • Reduce noise — especially for high ISO or low-light shots
  • Retouch scratches and tears — from old paper photos
  • Colorize black and white — historical images with plausible colors
  • Restoration — make faces recognizable again that have become unidentifiable due to aging
  • Change background — precise isolation thanks to AI segmentation

🟡 Engelmann Tools for Photo AI

This is exactly where our tool Photomizer comes in: AI-based image enhancement with one click – exposure, sharpness, colors, and noise are automatically optimized. For targeted enlargement, there's Photo BlowUp: The AI upscales images up to 1000 times without pixels becoming squares.

Both are practical: Restoring an old photo collection? Photomizer improves quality, BlowUp enlarges for printing. Both work locally on your PC – your family photos stay with you.

View Photomizer →  |  View Photo BlowUp →

Online vs. Local — When to Use Which?

  • Online AI (cloud-based) is often powerful, but: Your photos are uploaded. For family photos, that's not for everyone.
  • Local AI (on your PC) is privacy-friendly but requires software like Photomizer or special Stable Diffusion setups.
🎬

AI for Videos

Video AI is the hottest area in 2026 – and developing rapidly. Currently three important use cases:

1. Generate Videos from Text

Sora (OpenAI)

Generates photorealistic short videos from text descriptions. Results are now very close to real footage – including slow motion and complex camera movements.

Runway

Professional tool for video production. Text-to-video, image-to-video, video editing with AI. Used by filmmakers and content studios.

Luma Dream Machine

Slightly more accessible than Runway. Good motion representation, popular for social media content.

2. AI Avatars and Speakers

Synthesia

AI-generated speakers in 140+ languages. You input text, the AI creates a video with a photorealistic avatar. Popular for corporate videos and e-learning.

HeyGen

Similar to Synthesia. Strong in lip-sync and voice cloning. Have your own avatar speak foreign languages.

3. Enhance Existing Videos with AI

  • Upscale — Bring old VHS to 4K (Topaz Video AI is the standard here)
  • Stabilize — smooth out shaky footage
  • Automatically create subtitles — Whisper from OpenAI is free and very good
  • Remove / replace background — like greenscreen, without a greenscreen

📹 Practical Recommendation

For most home users, AI video is still more of a gimmick – the more exciting application is improving existing footage: digitizing old wedding videos on VHS and upscaling them with AI, then archiving them. This is the workflow that truly adds value.

📝

Write AI Texts

AI language models are the all-rounders among AI tools. Writing, translating, summarizing, programming, researching, brainstorming, analyzing – all in one tool.

The Big Three (and an Important Fourth)

ChatGPT (OpenAI)

The pioneer and market leader. Very broadly applicable, integrates image generation and computer use. The free version is good, the Pro version excellent.

Claude (Anthropic)

Particularly strong for long texts, analysis, and coding. Uses a very large context window (1 million tokens) – entire books can be processed in one prompt.

Gemini (Google)

Deeply integrated with Google services. Good for research, fact-checking, and multimodality. The "researcher" among the large models.

Llama / Mistral / DeepSeek (Open Source)

Freely available models, runnable locally. Now technically close to proprietary models – and full control over your data.

Which AI for What?

  • Brainstorming, creative texts: ChatGPT or Claude – both very strong
  • Analyze long documents: Claude (large context window)
  • Programming: Claude (Anthropic), GitHub Copilot, or ChatGPT
  • Research & fact-checking: Gemini or ChatGPT with web search
  • Translations: all three are very good, DeepL also remains a strong option
  • Privacy-sensitive content: local model with Ollama or LM Studio

Common Mistakes with AI Texts

  • Hallucinations: AI invents sources, facts, quotes. Always check, especially for important texts.
  • Generic style: Standard AI texts often sound smooth, interchangeable. Countermeasure: concrete examples, define your own style, iterate multiple times.
  • Outdated information: Training data has a cutoff date. Current data only available via web search function or manual verification.
  • Over-reliance: AI sounds convincing, even when it writes nonsense. Remain skeptical.

💡 Tip for better results: Give the AI a role and context. Instead of "Write me a text about X", rather say "You are a copywriter for senior magazines. Write a text about X in simple language, with everyday examples." Concrete specifications = concrete results.

🎨

Engelmann AI tools for your photos

Photomizer turns old snapshots into eye-catchers, Photo BlowUp enlarges losslessly for printing. Both work locally – your data stays with you.

More in the AI Compendium: