Generative AI
Un programma intensivo di 2 giorni: da zero al tuo primo video AI. Impara a generare immagini, animare personaggi e produrre contenuti visivi con l’intelligenza artificiale.
The workshop “Generative AI in 2 Days” offers a comprehensive immersion into the world of visual creation using artificial intelligence, through an intensive, hands-on 12-hour programme.
Designed for those who wish to learn AI generation techniques, the course is led by industry experts with years of experience in the fields of visual AI and film production. We will use professional cloud platforms and node-based visual canvases to generate photorealistic images and cinema-quality videos.
Objectives of the “Generative AI in 2 Days” masterclass
The masterclass aims to:
- A comprehensive overview of generative AI models for 2026
- Generate photorealistic images with complete control
- Reference image and character consistency across clips
- Creating AI videos with Kling and Seedance
- Apply motion transfer to real footage
- European AI Act and rights relating to generated images
Day 1 • Understanding AI. Generating images.
The day runs from 9.00 am to 5.30 pm and is divided into theoretical and practical sessions.
Generative AI Overview 2026
-
Images, video, audio and 3D: where we stand today in the AI landscape
-
Cloud platforms and node-based visual canvases for generating images and videos
-
Image models: Nano Banana Pro, Seedream, GPT Image 2, Flux, Wan 2.1
-
Video models: Kling 3.0, Seedance 2.0, LTX Video, Luma AI, Runway
-
Conversational AI systems such as prompt engineers: Claude, ChatGPT, Gemini
09:00 – 10:30
The AI Act & Image Rights
- The EU AI Regulation: obligations for content creators
- Copyright on generated images: who owns what
- The use of historical figures and public figures in AI
- The C2PA Standard and Content Credentials: source metadata
10:30 – 12:00
Lunch break
12:00 – 13:00
Generating Images Using Cloud Platforms and Node-Based Canvases
- Cloud platforms vs node-based visual canvases: how to choose the right tool
- The anatomy of the visual cue: subject + style + lighting + quality
- Negative prompts and quality parameters with Nano Banana Pro and GPT Image 2
- Practical session: 5 coherent variations on the same theme
13:00 – 14:30
Consistency between the Subject and the Reference Image
- Style reference, subject reference e composition reference
- IP adapters and cloud-based node canvases: preserving the character’s appearance and identity
- A practical workflow with Nano Banana Pro: using a ‘master’ image as the project’s foundation
- Practical session: the same subject in different styles and settings
14:30 – 16:00
Shots, Camera Work and Composition
- Angles: eye-level, bird’s eye, worm’s eye, dutch angle
- Focal lengths: 16mm wide-angle, 50mm standard, 85mm portrait, macro
- Depth of field: shallow DOF, deep focus, rack focus
- Inpainting: inserting subjects into real-world environments with full control
16:00 – 17:30
Day 2 • AI Video & Production.
The event runs from 9.00 am to 5.30 pm, featuring intensive practical sessions on video production
Introduction to Video AI
- Text-to-video, image-to-video, video-to-video: differences and uses
- Kling 3.0: subject consistency and natural body movements
- Seedance 2.0: excellent for motion transfer and visual consistency
- Camera control nel prompt: dolly in, pan, static, crane shot
09:00 – 10:30
Motion Prompt
- Structure: subject action + camera movement + atmosphere
- First frame / Last frame: check the start and end of the video
- Use Claude to write motion prompts in seconds
- Practical session: animating the images created on day 1
10:30 – 12:00
Lunch break
12:00 – 13:00
Motion Transfer from Real Footage
- Seedance motion transfer: your movements are replicated by the AI character
- How to shoot an effective reference video with your smartphone
- Practical session: filming in the classroom and applying it to your character
- Comparison between pure motion prompts and motion transfer from footage
13:00 – 14:30
Structuring a 30–60-second video
- 5–8 clips, each 4–8 seconds long: plan before creating
- Narrative arc: beginning, development, climax, conclusion
- Visual consistency: the same colour palette, the same lighting, and the same grain across all clips
14:30 – 15:30
Upscale, Assembly and Delivery
- Topaz Video AI: upgrading clips to 4K quality for final delivery
- Editing with CapCut (free), DaVinci Resolve or Premiere Pro
- Royalty-free AI music: Suno, Udio — how to incorporate it into your video
- Presentation of work, open Q&A on the creative AI profession
15:30 – 17:30
Course technology stack
IMAGE GENERATION
Nano Banana Pro
GPT Image 2
Seedream
Flux · Wan 2.1
Canvas cloud a nodi
Topaz (upscale)
VIDEO GENERATION
Kling 3.0
Seedance 2.0
LTX Video
Runway Gen-3
Luma AI
PROMPT & AI
Claude (Anthropic)
ChatGPT (OpenAI)
Gemini (Google)
Prompt engineering
Style consistency
WORKFLOW & POST
DaVinci Resolve
CapCut (free)
Adobe Premiere
Suno / Udio
WeTransfer
Teacher
Paolo Aralla
Founder & CEO · Lecturer in Generative AI · Universities of Bergamo and Turin
The team leads the workshop, offering cutting-edge training in the field of visual AI. Drawing on advanced technologies and extensive experience, our instructors deliver high-quality training, exploring the latest innovations in the sector. The team has contributed to successful film and TV series productions, working closely with leading Italian and international VFX supervisors.
