July 1, 2026
Pilot in your project: a chat that picks the model and runs the generation
Every Givon AI project now has a third surface — Pilot. It's a normal chat: ask for advice or just talk. Or ask it to create something, and it picks a model, writes the prompt, shows the exact price, and runs the generation only after you approve it.
Every Givon AI project now has a third tab — next to Materials and Editor, there's now Pilot. It's a normal chat: ask for advice, talk through an idea, ask it to break down a brief — it replies like any AI chat, without generating anything. But ask it to create something, and it switches gears.
On an image or video brief, it picks a model from the live catalog, writes an engineering prompt, and shows a proposal card: model, settings, a prompt excerpt, and the exact cost in tokens — before anything runs. Generation only starts after an explicit Approve. Just want a ready prompt? Ask for it directly — Pilot returns it as text in a code block, with no run and no generation charge.
The whole path stays visible in the chat: launch, a "Generating…" status, the finished result — previewed right there, with the asset already sitting in Materials. If something fails, the chat shows the error and offers "Fix & retry": Pilot diagnoses what went wrong, rewrites the prompt, and proposes a new run instead of a blind retry.
Pilot also has memory: describe your project's style, brand, characters, or rules once, and it keeps that in mind in every chat. You can run several chats per project in parallel, with search and renaming — just like the AI chats you already use.
On this page
What changed
Just a chat
Ask anything: ideas, copy, advice, feedback on a brief — a normal conversation, no generation involved.
A proposal with a price
On an image or video brief, Pilot picks a model and settings, writes the prompt, and shows the exact cost — it only runs after you hit Approve.
A prompt on request
Ask for "just the prompt" and get ready text in a code block — no run, no generation charge.
The whole path is visible
Launch, generation status, the finished result, or an error — right in the chat; the result lands straight in your project's Materials.
Fixes on failure
If a generation fails, Pilot offers "Fix & retry": it diagnoses the cause, rewrites the prompt, and proposes a new run.
Remembers your project
Save style, brand, characters, and rules to your project's memory — Pilot applies them on every turn.
Conversation
A normal chat — and a generation agent — in one window
Open Pilot tab in a project and you'll see your chat list on the left and a message box at the bottom. Ask anything — brainstorm ideas for a clip, explain a task, check which model fits a vertical video best — Pilot replies like a normal AI chat, without running or charging for a generation.
The difference starts when you ask it to create something. Pilot reads the brief, checks project materials or the model catalog if it needs to, writes a prompt through the same engine that powers "Help with the prompt" in the generation bar — and shows you a proposal instead of generating blind.
Proposal
A proposal card: model, settings, and price — before you run it
When Pilot is ready to propose a generation, a PROPOSAL card appears in the chat: the model and key settings, a prompt excerpt (expandable to the full text), and the exact cost in tokens — not an estimate range. If you have a Membership discount, the card shows the strikethrough base price next to the discounted total.
From there it's Approve or Reject. Nothing runs and nothing is charged until you hit Approve — even though Pilot already spent time picking the model and drafting the prompt for that turn.
The price is an estimate
Just like in the editor, the shown price is a preliminary quote; the exact amount is locked in at launch. If a proposal goes stale or the price drifts noticeably, Pilot asks you to refresh the request instead of charging the old number.
Lifecycle
From launch to result — without leaving the chat
After you approve, the generation runs the same way it would from the editor: the same queue, the same billing, the same engine. Meanwhile the chat shows a live status — "Generating…" — and once the asset is ready, an inline preview appears with a link straight to Materials.
If a provider rejects the generation or something breaks, Pilot doesn't just say it failed. It offers "Fix & retry": it works out what went wrong, rewrites the prompt (removing whatever might have triggered a block, or simplifying the request), and proposes a fresh card — with a new approval, not an automatic retry on your dime.
Control
Your choice of approval mode and Pilot model
By default Pilot runs in Confirm mode: every launch is a card with a price and an explicit Approve. If pausing on every detail isn't what you want, switch to Auto (runs without pausing, up to a daily budget) or Hybrid (auto-runs cheap generations, still shows a card for expensive ones).
You can also pick which model drives the conversation itself: Opus 4.8 for complex briefs and fine edits, Sonnet 5 for most tasks, Haiku 4.5 for quick questions where speed matters. Each option has its own per-turn cost, shown right in the model picker before you confirm it.
Memory
Pilot remembers your project — and you can run several chats
In "Project context" you can save notes by category — Rules, Brand, Personality, Style, Habits, General. For example: "the hero is a ginger cat named Mars" or "no neon." Pilot factors these notes into every turn, so you don't have to re-explain context each time.
You can keep more than one chat per project — "New chat" starts another thread with its own title (auto-generated from the first message, but renameable). The chat list is searchable right there on the left, and a link to a specific chat is preserved in the URL.
What's next
What's coming next for Pilot
Pilot will keep growing. Coming up next:
Photos in chat
Attach an image right in the conversation and ask for edits without leaving the chat.
Skill management
Turn Pilot's knowledge packs on and off by hand, for a specific task.
This is a plan, not a date
Timing isn't fixed. We'll cover each step separately — keep an eye on the Updates section.
Questions
Frequently asked questions about Pilot
- Is this the same as "Help with the prompt" in the generation bar?
- No. "Help with the prompt" improves a prompt for the model and mode you've already picked. Pilot is a separate chat: it can pick the model itself, explain why, write the prompt, and — with your approval — run the generation, or just answer a question with no generation at all.
- Do I get charged tokens just for asking for advice?
- Some messages each month aren't charged; beyond that, every turn costs a small flat amount that depends on which Pilot model you picked. Generation itself is only charged once you approve a specific proposal with a shown price.
- Can I send Pilot a photo?
- Not yet — that's still in development. Right now Pilot works with text and with materials already uploaded to the project; attaching photos directly in chat is coming later.
- What happens if a generation fails?
- Pilot shows the error right in the chat, explains the cause, and offers "Fix & retry" — it rewrites the prompt and proposes a new run instead of a blind retry.
- Can Pilot generate music, voiceover, or avatars?
- Right now the chat can propose image and video generations. Other generation types through Pilot are coming separately.
- Where are my chats with Pilot stored?
- Each project has its own chat list, with search and renaming. You can run several chats per project in parallel and switch between them.