Tired of ads? Enjoy an ad-free experience by signing up.
👩‍🍳 How we use AI at Tech in Asia, thoughtfully and responsibly.
🧔‍♂️ A friendly human may check it before it goes live. More news here

US AI startup General Intuition raises $133.7m seed funding

General Intuition, an AI research startup, has raised US$133.7 million in seed funding to develop agents with spatial-temporal reasoning using video game clips.

The round was led by Khosla Ventures and General Catalyst, with participation from Raine.

General Intuition is a spinout from Medal, a platform where users upload and share video game clips.

The company is using Medal’s dataset—2 billion videos annually from 10 million monthly active users—to train AI agents that can understand how objects move through space and time.

Initial applications include gaming and powering search-and-rescue drones.

🔗 Source: TechCrunch

🧠 Food for thought

Implications, context, and why it matters.

Medal’s broad user-generated content (UGC) license allows AI training, but underlying game intellectual property (IP) may complicate authorization

  • Medal’s terms give it a broad license to use, modify, plus monetize uploads including for AI training and third-party sharing 1. That does not answer whether it can use the copyrighted footage in them, including General Intuition’s work 1.
  • Publishers usually own copyrights in game content and assets. Gameplay clips can touch rights an uploader may not fully control 2.
  • Medal tells users to post material they own or have permission to share. It relies on notice-and-takedown and Digital Millennium Copyright Act (DMCA) processes instead of checking licenses upfront 1.
  • This puts General Intuition in a gray area. Publishers could claim copyrighted material is used without permission, similar to author lawsuits over AI models trained on pirated books 3.

AI tool providers serving game developers need rights-clearing services for training data

  • Game makers using AI tools face tighter rules. On Steam, teams must disclose AI use plus prove training-data rights. Games with unlicensed material can be rejected 2.
  • Many publishers are moving from one-time training deals to usage-based grounding licenses, which tie model behavior to specific data sources. These deals pay per query or crawl and create recurring revenue for brokers 4.
  • A third-party service could map AI policies, negotiate blanket licenses for gameplay footage, and provide compliance docs to help AI labs plus teams that must prove legal training. Rights-clearing means getting IP owner permissions and keeping auditable records 2.

Stay ahead in Asia’s tech landscape

You've reached your 2 free content limit for the month. Sign up for free to read the full story.

🏄 For casual readers / 👶 Free

Basic

US$0

Free forever

Get instant access to this article and more every month

0 premium content

Unlimited news briefs

5

5 articles

Ad-free reading experience

Just US$0 per day

⌛Sign up in 20s. No payment details needed.

📖 For learners / 👍 Starter

Lite

US$4.92/month

Billed annually at US$59/year

Get instant access to this article and more every month

4

4 premium content

Unlimited news briefs & articles

Ad-free reading experience

Just US$0.17 per day

Cancel anytime

Our subscriber community includes professionals from these companies:

Stay updated on the go with our mobile app.

Get latest insights with smoother, more personalized experience through TIA mobile app.