Uploaded July 2025 | Updated September 2026, 3 weeks ago

---

This short film is based on the report AI 2027: ai-2027.com

Shrinking AGI timelines: a review of expert forecasts 80000hours.org/2025/03/when-do-experts-expect-agi-to-arrive
The case for AGI by 2030 80000hours.org/agi/guide/when-will-agi-arrive

This video was supported by the Future of Life Institute’s Digital Media Accelerator: futureoflife.org/project/digital-media-accelerator

This video is probably the hardest on I’ve worked on. It took me a month of full time work, plus I got a lot of help from a friend with the editing. It took a lot more work than my usual video, ironically enough. These were the AI tools used in the making of this video:

Main character’s video and audio - HeyGen + ElevenLabs
B-roll video - Veo 3 (thank you Google Deepmind for free access)
Other characters talking - Veo 3 (and sometimes ElevenLabs as well)
Scripting - Claude 4 Sonnet/ Opus
Character design and frames - Flux Kontext in OpenArt
Agent 3 and 4 design and stills etc - ChatGPT o3 image model + Canva
Thumbnail - ChatGPT o3