Can AI Learn to Skateboard? Testing AI's Skateboarding Skills
Braille SkateboardingJanuary 10, 202620 min2,632 views
12 connectionsΒ·16 entities in this videoβAI's Attempt at Skateboarding
- π€ The video explores the capabilities of AI by attempting to generate skateboarding scenes, starting with basic tricks like ollies.
- β οΈ Initial AI-generated clips show skaters with incorrect board orientation, wheels rolling sideways, and failed ollie attempts.
- π‘ The creator provides detailed feedback, emphasizing fundamental skateboarding mechanics like wheel direction and proper trick execution.
Refining AI Skateboarding Prompts
- π― Subsequent attempts focus on more specific actions, such as riding a skateboard down a street or performing a kickflip.
- π Despite detailed instructions, AI struggles with realistic physics, trick execution (e.g., kickflips, shove-its), and consistent character actions.
- π§ The creator even uses ChatGPT to generate a better prompt, but the results remain flawed, highlighting AI's difficulty with nuanced physical actions.
Iconic Skateboarding Scene Recreation
- π₯ The experiment extends to recreating famous Braille skateboarding scenes, like dropping in on a glass skateboard.
- π§ AI fails to accurately depict the scene, altering the skateboard's material mid-action, misinterpreting the drop-in, and showing incorrect physics.
- π Another attempt to recreate a Christmas tree bungee jump scene also results in nonsensical physics and character actions.
Conclusion on AI and Skateboarding
- π« The creator concludes that AI, despite its advancements, cannot currently replicate the complexities and nuances of skateboarding.
- π The video emphasizes the value of human learning, progression, and the fun of skateboarding, suggesting AI is not a threat to human skateboarders.
- π¬ Viewers are encouraged to share their thoughts on AI's capabilities and limitations in the comments.
Knowledge graph16 entities Β· 12 connections
How they connect
An interactive map of every person, idea, and reference from this conversation. Hover to trace connections, click to explore.
Hover Β· drag to explore
16 entities
Chapters8 moments
Key Moments
Transcript75 segments
Full Transcript
Topics12 themes
Whatβs Discussed
Artificial IntelligenceSkateboardingAI GenerationOllieKickflipShove-itPhysics SimulationPrompt EngineeringChatGPTVideo GenerationSkateboarding TricksAI Limitations
Smart Objects16 Β· 12 links
ProductsΒ· 5
PersonΒ· 1
CompaniesΒ· 2
ConceptsΒ· 4
MediasΒ· 3
LocationΒ· 1