The studio behind the Avengers movies went to Tokyo this week and said something I’ve been waiting for somebody with a big budget to say.
At TIFFCOM, the content market of the Tokyo International Film Festival, AGBO’s Chris Brearton and Dominic Hughes gave a keynote on AI and storytelling. AGBO is the Russo brothers’ shop. Their take: AI is an empowerment. It handles “all the annoying things that keep you from enjoying yourself,” and it will matter most in process, not in replacing the people who make the stuff. Hughes, their chief scientific officer, said machines are good at “regurgitating what has been done in the past,” not at real novelty. Ask one to write its own Shakespeare play and “you’d have the sense that it was written by a 12-year-old.”
Surprise. A tech guy saying the tech isn’t the artist.
Then he said the part that stuck with me. AI, Hughes said, “is good at critiquing.” It can tell you whether a play somebody found in an attic might be Shakespeare’s.
Fine. I believe it. And that’s exactly why we don’t let it near reviews.
Look at the whole picture. Hughes also said AI can generate endless amounts of mediocre content, “but consumers can’t possibly absorb it,” and the only answer is to raise the bar on quality. Agreed. So who raises the bar? Not the thing that made the mediocre content in the first place. People do. People who read your chapter and felt something, or didn’t, and can tell you why.
That’s the little world I care about, and Circle is being built for exactly that.
Here’s how it works. Members request assignments. Reviews are structured and prompted, submitted right on the site, and they walk the reviewer through different lenses. They push for real feelings, not polite noises. Writers can revise and resubmit. Reviewing earns credit, and you can spend that credit getting eyes on your own work or put it toward someone else’s.
On AI, Circle doesn’t try to make the law. No complicated formulas for what’s allowed. Before your first submission you’re asked one time to be honest about if and how you use AI. It’s part of the membership agreement. Honesty, that’s what we ask. It’s your word and your reputation. How you feel about AI in your own writing or production is your call.
The one place AI isn’t allowed is reviewing someone else’s work. We are not interested in how AI feels about it. You can go get an AI opinion from an AI chatbot any time you want. What we want from a review is honest, informed human feedback.
And maybe AI gets good enough to beat all of us at “best” someday. Hughes thinks the Beethovens of the world can’t be replicated. I don’t know. Doesn’t matter. Circle can help someone who wants to be the best, but that’s not who it was built for. It’s for writers who want to be their best. The ones who write for the love of it, because they have to.
Now the honest part. Circle is new. There’s almost nobody in it yet. The bones are in place. What we need are founding members who want to get to work and who’ll have a real say in how this thing grows and how the vision gets built. If you’re looking for a crowd, come back later. If you want to build, come now.
If that’s you, join us and earn your satchel and your desk.
Registration is open at gapoz.com. No fees. Membership to the Circle is limited and subject to approval. Satchels are earned, not given away. Come meet people, make some friends, and grow your craft together.
Your name. Your word. Your work.
Also on the beat
- Publishers file an amicus brief urging the court to reject OpenAI’s fair-use defense in the Authors Guild case, citing ChatGPT “sequels” to George R.R. Martin. Publishers Weekly, Oct 29
- Grok Imagine’s seven-minute AI Odyssey episode gets slated by critics, with The Guardian calling it “meaningless as a laptop screensaver.” The Guardian, Oct 27
- The New York Review of Books looks at the mistrust behind the Thélyson Orélien affair. NYRB, Oct 29
- Authors accuse publishers of “feeding at both sides of the trough”: suing AI companies while using AI in-house. Tomorrow’s Publisher, Oct 29
- SAG-AFTRA’s contract makes studios prove “significant additional value” before using a synthetic performer, but nobody has defined it yet. AI Lately, Oct 29
How this was made
AI assisted research and a first draft of this Circle View. Richard Crossley reviewed, edited, and is accountable for the judgment and the final take.
