Generative AI now a normal ideation step

Design teams are running tools like Midjourney, ChatGPT and Vizcom through early concepting as a routine speed-up, not a gimmick. The skill gap now is knowing how to use them, not whether to.

Studios are casually running Midjourney, ChatGPT and tools like Vizcom through early concepting — not as a gimmick, but as a faster way to generate visual options before committing real design time. This kind of experimentation frames generative AI as a normal part of ideation and prototyping stages, not bleeding-edge novelty anymore. For junior designers, the risk isn't 'AI will replace you' — it's showing up to a job not knowing how to use these tools as a first-draft generator. You don't need to be an expert; you need one AI tool in your process you can talk about with specifics, not buzzwords, when someone asks about your workflow.

Related in AI workflows

Adoption

MCP goes mainstream across creative tools

At SIGGRAPH 2026, NVIDIA highlighted how major creative apps are now exposing Model Context Protocol (MCP) connections, letting AI agents work inside design and production tools while creators keep control. This follows Figma's own MCP server rollout earlier in the year.

blogs.nvidia.com10 Aug

Established

AI now handles the boring bits of creative pipelines

Autodesk's SIGGRAPH 2024 tools show AI embedded in production pipelines, not just demos. Scheduling and rigging tasks get automated so artists spend more time on judgement calls.

adsknews.autodesk.com29 Jul

Adoption

Big companies are building in-house AI design flows

UX Collective rounded up how teams like Atlassian are wiring AI into specific steps of their design-to-prototype process, not replacing it wholesale. The pattern: narrow AI tasks get adopted, broad 'do my job' AI doesn't.

uxdesign.cc15 Aug