In a rapidly evolving business landscape, the integration of Artificial Intelligence (AI) has moved from being a novelty to an essential component of daily operations. In his keynote at Affiliate World, Rory Flynn, founder of Systematic AI, provided insights into how his company is leading this transformation.
AI is no longer just a buzzword; it’s a critical tool reshaping the way businesses operate. Flynn emphasizes that while many are already using AI, the real challenge is thinking both big and small—injecting AI into every possible task and workflow. From automating mundane tasks like call note entries into a CRM, to creating comprehensive automated design systems, the possibilities are vast.
“For us, AI is no longer novel. It’s essential. It needs to get bigger and we need to figure out how to do that.“
The five stages of AI adoption
Flynn outlines a clear progression for AI adoption in businesses:
- Discovery: At this initial stage, businesses explore AI tools and recognize their potential to address various challenges.
- Adoption: AI begins to be integrated into specific business functions, tackling individual problems and improving targeted processes.
- Fragmentation: As AI usage becomes widespread, a lack of unified strategy leads to inefficiencies and disjointed applications.
- Systematization: Businesses establish standardized operating procedures (SOPs) to ensure consistent and effective AI usage.
- Automation: The final stage involves refining and fully automating SOPs, allowing AI to operate independently and seamlessly within business workflows.
Most companies are currently navigating between the adoption and systematization stages, working towards a more integrated and efficient use of AI.
“I think people utilizing AI right now, they’re just not thinking big enough, and they’re not thinking small enough. What that means is AI can essentially be injected into every workflow, every task, every little micro task, but it just takes some thought on how to use it.”
The need for speed and precision
Flynn’s team at SuperSight, a prominent design agency, has developed an automated system that delivers final products in just 24 seconds from a brief. This system, powered by custom-trained models, eliminates the traditional back-and-forth of client revisions and streamlines the creative process. The result is significantly faster delivery times and increased precision—essential for meeting the high-volume demands of modern brands.
Components of an AI-powered design system
Flynn’s automated design system hinges on four crucial components:
- Knowledge graphs: These serve as comprehensive repositories for training agents and maintaining quality control.
- Agents: Specialized entities within the system that perform specific tasks, such as image creation, copywriting, and template design.
- Integrations: Tools that connect agents to ensure smooth communication and task execution.
- Coding: Ensures effective communication between components, converting outputs into standardized formats like JSON.
Together, these components enable the system to process human inputs (e.g., design briefs), execute tasks, adhere to brand guidelines, and generate fully designed products in Figma within seconds.
Impressive ROI and future growth
The results from Flynn’s system are striking. Deployed in over 1,000 projects, the system has achieved a 200% faster completion rate, a 50-85% reduction in team hours, and over $1.4 billion in savings for clients. Flynn emphasizes that this is just the beginning; the system is poised for further growth and enhancement in the coming months.
Practical applications: Asset hacking with AI
A notable example of Systematic AI’s approach is the asset hacking process demonstrated at Affiliate World Bangkok. By utilizing tools like Midjourney and ChatGPT, Flynn’s team successfully reverse-engineered a single image into a thousand variations. This technique, which involves converting images into text-based prompts, enables rapid and scalable content creation adaptable to various brand requirements.
Building with a plan
Flynn provides a framework for businesses aiming to build AI-driven systems. The process starts with defining the end result—such as fully designed ads—and breaking down tasks into detailed data components. These components are then reassembled using AI, ensuring that integration is value-driven rather than a patchwork solution.
Key considerations for AI integration
- Data breakdown: Dissecting components like images, copy, and templates into manageable data sets is crucial for effective AI integration.
- Role of agents: Specialized agents should focus on specific tasks (e.g., image generation, copywriting) and be customized to meet precise needs.
- Integration and communication: Utilize tools like BlankChain for seamless communication between agents and standardize data formats with JSON.
- Importance of data quality: High-quality input data directly impacts output quality. Investing in well-curated datasets is essential for optimal performance.
- Scalability and flexibility: Tailoring agents to specific tasks allows for efficient handling of multiple clients or brands and fosters creativity.
Future directions
Looking ahead, AI tools are rapidly evolving. Businesses should stay updated with advancements while focusing on mastering a few reliable tools. The goal is not to achieve perfection in every output but to rapidly produce drafts that can be refined with human creativity. Automation and ongoing improvements will continue to enhance the efficiency and effectiveness of AI-driven processes.
Flynn’s keynote underscores the transformative potential of AI when applied thoughtfully and systematically. By adopting a strategic approach to AI integration, businesses can achieve speed, precision, and cost savings, positioning themselves at the forefront of the generative future. As AI continues to evolve, companies that adapt quickly and strategically will be best positioned to thrive in this landscape.
