Recent signals from the games industry point to a broader shift: companies are preparing for a future where people and AI create products together. For business leaders in Greater Barcelona, this is not only a technology story. It is a management question about workflow design, governance, intellectual property, quality control, and team capability. The real issue is not whether AI will participate in product and content development, but how companies will structure that participation responsibly.
What AI co-creation actually means for business
AI co-creation does not mean handing product decisions to a model. In practice, it means using AI to support defined parts of the work: idea generation, draft production, research summarisation, asset variation, documentation, testing support, or process acceleration. Human teams still set priorities, approve outputs, protect brand standards, and decide what goes live.
For SMEs, this matters because AI changes the cost and speed profile of knowledge work. Tasks that were previously too slow, too expensive, or too manual can now be redesigned. That creates opportunities, but also operational risk if companies adopt tools before defining rules.
Why this matters beyond gaming and media
The headline may come from a creative industry, but the underlying model applies far more widely. Software companies can use AI to accelerate specifications and testing. Manufacturers can use it in technical documentation and internal knowledge access. Service businesses can apply it to proposals, reporting, training materials, and customer support workflows. Marketing teams can use it for campaign ideation and content production, provided review controls are in place.
The strategic point is simple: AI is becoming part of the production environment, not just an experimental side tool. Businesses that treat it as a standalone app often get fragmented adoption, inconsistent output, and avoidable compliance concerns.
The management questions leaders need to answer first
Before scaling AI-assisted creation, leadership teams should answer five practical questions. First, which workflows are suitable for AI support, and which are not? Second, what level of human review is mandatory before release or publication? Third, what data can be used safely in prompts and models? Fourth, who owns the process and measures output quality? Fifth, how will teams document decisions, exceptions, and risks?
These are governance questions, not just IT questions. If they remain unresolved, companies usually end up with scattered experimentation, inconsistent quality, and uncertainty around accountability.
A workable operating model for AI-assisted development
A useful starting model is to separate work into three layers. The first layer is low-risk assistance, such as summarising internal material, generating first drafts, or proposing alternatives. The second layer is supervised production, where AI contributes to customer-facing or product-related outputs that require structured human review. The third layer is restricted activity, where AI use is limited or prohibited because of confidentiality, regulatory, contractual, or brand risk.
Once these layers are defined, teams can assign approval rules, tool access, documentation requirements, and escalation paths. This helps companies move from informal experimentation to repeatable execution.
For many SMEs in Greater Barcelona, this is the most practical path: start with a narrow set of high-friction workflows, define guardrails, and scale only where quality and accountability are clear.
How to build governance without slowing the business down
Good governance should enable adoption, not block it. A lightweight framework is often enough at the start. Define approved use cases. Create simple prompt and review guidelines. Set rules for confidential information. Clarify when AI-generated material must be checked by legal, product, or brand owners. Keep a record of tools in use and who is responsible for each workflow.
This is also where digital strategy becomes relevant. AI co-creation should support business goals, operating priorities, and capability building. It should not become a disconnected tool rollout driven only by short-term curiosity.
What business leaders should do next
Start with an audit of current knowledge-heavy workflows. Identify where teams spend time on repetitive drafting, internal search, formatting, coordination, or first-pass analysis. Choose two or three use cases with clear value and manageable risk. Define the expected output, review process, owner, and success criteria before introducing tools.
Then run a controlled pilot. Measure adoption, cycle time, rework, and quality issues. Document where human intervention remains essential. If the pilot works, formalise the process and train the relevant teams. If it does not, adjust the workflow before expanding.
The companies that benefit most from AI co-creation will not be the ones with the most tools. They will be the ones that make clear decisions about where AI fits, where it does not, and how people remain accountable for outcomes.