Less is More in AI: The Psychology Behind Prompts that Save Tokens and Optimize Design

Interaction with generative artificial intelligences has become an indispensable tool in the arsenal of designers and creative professionals. However, the effectiveness of this collaboration lies not only in the AI's capability, but fundamentally in the quality of human communication. The premise that "less is more" in AI prompts is not merely a cliché; it is a principle rooted in resource economy and, more deeply, in the cognitive psychology that governs how we process and interpret information.

The Economy of Tokens and AI Efficiency

To understand the importance of conciseness, it is fundamental to grasp the concept of "tokens". Language models, such as those used in generative AIs, process information in units called tokens. A token can be a word, part of a word, or even a character. Each interaction with the AI consumes a number of tokens, both in the input (the prompt) and in the output (the generated response).

This token economy has direct implications:

• Cost: Many AI models operate with a cost per token. Longer and redundant prompts result in higher costs.
• Processing Speed: Extensive prompts require more time for the AI to process, impacting workflow agility.
• Context Window: AIs have a limited "context window," which is the maximum number of tokens they can process in a single interaction. Excessively long prompts can consume this window, limiting the AI's ability to consider crucial information or generate more elaborate and contextually rich responses.

Conciseness, therefore, is not just a matter of financial or time economy; it is a strategy to maximize the AI's processing capacity, ensuring that the limited space of the context window is filled with truly relevant and directional information.

Cognitive Principles in Interacting with Language Models

The way we formulate prompts largely reflects our own cognitive processes. When interacting with an AI, we are, in a way, "training" or "directing" its "reasoning" for a specific task. Cognitive psychology offers valuable insights on how to optimize this communication.

Cognitive Load and AI Focus

In human psychology, cognitive load refers to the mental effort required to process information. Although AIs do not possess a "mind" in the human sense, we can analogously think of a "processing load". A verbose prompt, full of irrelevant information, redundancies, or ambiguous instructions, increases this load. The AI needs to "filter" the noise to find the signal, which can lead to:

• Focus Deviation: The AI may get stuck on secondary details, losing sight of the prompt's main objective.
• Generic Results: To avoid errors amidst ambiguity, the AI may opt for safer and less creative or specific responses.
• Hallucinations: In extreme cases, information overload can lead the AI to "invent" data or connections that do not exist in the prompt.

Concise and direct prompts reduce this load, allowing the AI to focus its computational resources on essential information and the stated objective, resulting in more precise and aligned outputs.

Clarity and Ambiguitiy: The AI's Perspective

Humans are masters of inference. We can fill in gaps, understand subtexts, and interpret intentions even with imperfect communication. AIs, on the other hand, are fundamentally literal. They process what is explicitly provided.

• The Curse of Knowledge: Often, when writing prompts, we assume that the AI possesses the same context or implicit knowledge as we do. This "curse of knowledge" leads us to omit crucial details or to use language that, to us, is clear, but to the AI, is ambiguous.
• The Need for Specificity: Clarity for an AI means specificity. Instead of "Create something beautiful," which is subjective, an effective prompt would be "Generate a minimalist icon for a meditation app, using shades of blue and green, with organic shapes and no text".

Conciseness, in this context, does not mean a lack of detail, but rather the elimination of unnecessary words that do not add value to the instruction, while maintaining specificity and clarity.

Working Memory and Context

The AI's context window can be compared to human working memory, which has a limited capacity to actively retain and manipulate information. The more concise the prompt, the more "space" in the AI's working memory is freed up to process the most recent and relevant information, or to maintain a longer and more coherent conversation history. This is crucial in iterative interactions, where the AI needs to remember what was previously discussed to build upon it.

Strategies for Optimized Prompts in Design

Applying these cognitive principles in design practice with AI involves adopting specific strategies:

1. Define the Main Objective: Before writing, be clear about what you want to achieve. What is the desired final outcome? This helps filter out irrelevant information.
2. Be Specific, But Not Redundant: Provide crucial details (color, style, format, function), but avoid repeating information or using superfluous adjectives that do not alter the meaning.
   • Ineffective example: "Create a very, very beautiful and super modern logo design for a technology company that is innovative and wants something that looks futuristic and cool."
   • Effective example: "Generate a minimalist and futuristic logo for a technology startup. Use sans-serif typography and a blue and silver color palette."
3. Utilize Structures and Constraints: Guiding the AI with clear formats (lists, bullet points, short paragraphs) and explicit constraints (e.g., "no text", "icons only", "SVG format") helps direct the output.
4. Prioritize Key Information: Place the most important information at the beginning of the prompt. This ensures the AI processes them with greater weight.
5. Iterate and Refine: Rarely will the first prompt be perfect. Start with a concise version and add details or adjust the language in subsequent interactions, learning how the AI responds to different formulations.
6. Avoid Unnecessary Jargon: While technical design terms are useful, avoid overly complex or ambiguous language that might confuse the AI. Use the most direct language possible to describe what you need.

Tangible Benefits for Designers

Mastery in the art of concise and clear prompts translates into concrete benefits for the design process:

• More Accurate Results: The AI better understands the intention, generating outputs that align more closely with expectations.
• Greater Efficiency: Less time spent formulating long prompts and fewer iterations to correct misunderstandings.
• Cost Reduction: Token economy translates into financial savings, especially in large-scale projects.
• Workflow Optimization: A more fluid and predictable interaction process with the AI.
• Enhanced Creative Control: By precisely directing the AI, the designer maintains greater control over the creative direction, using the AI as an extension of their own vision, rather than as a random generator.

Mastering the art of concise and clear prompts is more than a technique; it is a deep understanding of how artificial intelligence processes information, mirroring principles of human cognition. For designers, this means not only saving resources but also elevating the quality and relevance of the generated outputs, transforming AI into a truly powerful and efficient collaboration. Conciseness, far from being a limitation, is the key to unlocking the maximum potential of AI in design.