Questions we hear often

Is VisualForge AI a lifestyle brand (because of .life), an NFT art project, or a course — and do you guarantee the visuals will perform or go viral?

No. We are an AI brand-visual studio that builds on-brand visual systems and generates content visuals at scale for client brands. The .life TLD is branding only — NOT a lifestyle, wellness, or coaching brand; we are NOT an NFT or crypto art project, NOT a course, and NOT a self-serve app. We train house styles on client-owned or licensed references, respect IP and likeness rights, keep a human art director in the loop, and disclose AI use where required. We do not guarantee reach, engagement, virality, or sales; outcomes depend on the brief, budget, and how the imagery is used.

How do engagements work — project or retainer?

Projects suit launches, visual-system builds, and house-style model training — fixed scope with a defined delivery schedule. Retainers suit ongoing content visual production where you need batch generation weekly or monthly with a consistent art-direction contact. We quote in CAD after a brief review call; typical onboarding takes 3–5 business days for retainer setup and 1–2 weeks for project kickoff depending on reference-library preparation.

What are typical CAD budgets?

Content visual batches from C$3,500; visual strategy from C$8,000; house-style training from C$12,000; full campaign visual worlds from C$15,000; retainer production from C$4,500/month. Final quotes depend on volume, revision rounds, governance requirements, and timeline. We provide indicative ranges before any commitment.

Which models and tools do you use?

We work with cloud-hosted and open-weight diffusion models depending on client governance, data residency, and quality requirements. Workflows include text-to-image, img2img reference generation, LoRA fine-tuning, inpainting, outpainting, upscaling, and composition control. Tool choice is scoped per engagement — we do not resell model access or run a prompt marketplace.

Who owns the imagery, prompts, and trained styles?

Ownership and licensing terms are defined in your statement of work. Clients typically receive usage rights to deliverables for agreed channels and durations. Trained model weights, prompt libraries, and style documentation may be licensed to the client, retained by the studio for the retainer term, or jointly governed — we document this clearly before training begins.

How are references and training data licensed?

We require client-owned assets, properly licensed stock, or references with documented clearance. We do not scrape copyrighted material or train on likenesses without consent. Your legal team receives a reference audit summary as part of delivery ops.

How do you handle likeness and consent?

Real individuals appear in generated imagery only with documented consent and appropriate release forms. We flag unintended likeness matches during human QA and remove or revise outputs before delivery. We do not create non-consensual or deceptive deepfake imagery.

How is AI use disclosed?

We advise clients on disclosure language for platforms, packaging, and brand guidelines where policy or jurisdiction requires it. Our deliverables include generation notes suitable for legal and compliance review.

How do you catch mistakes and artefacts?

Every batch passes human-in-the-loop review. Retouchers fix hands, text, colour drift, and hallucinated details. Art directors check brand consistency against style guardrails. Generative-AI output is probabilistic — we correct and finish by hand but do not guarantee pixel-perfect style matches on every frame.

What about revisions and data handling under PIPEDA?

Revision rounds are scoped in your statement of work — typically two rounds per batch for retainer clients. Personal information collected through our contact form is used only to respond to enquiries, stored securely in Canada where possible, and handled per our Privacy Policy and PIPEDA. We do not buy or sell personal data or personal images.

What do you NOT do?

We do not guarantee virality, reach, or sales. We do not create deepfakes, non-consensual likeness imagery, or NSFW content. We do not sell courses, NFT drops, stock images, or prompt/model marketplaces. We do not offer get-rich-with-AI schemes or impersonate real people or brands without consent.