Generative AI Guide

The AI Fluency Framework was developed by Rick Dakan and Joseph Feller, and is a practical tool for understanding the competencies that can help users use and make decisions about use and output of Gen AI and other artificial intelligence. 

Core AI Fluency Competencies: 4Ds Framework

Delegation: Creative vision and selection of the right AI tools and techniques to realize that vision.
Description: Effectively describing a vision and/or tasks to prompt useful AI behaviours and outputs. 
Discernment: Accurately assessing the usefulness of AI outputs
Diligence: Taking responsibility and vouching for final products created using AI 

Delegation - Creative vision and selection of the right AI tools and techniques to realize that vision.
Delegation refers to the ability to identify when and how to use AI tools and modalities effectively in creative and problem-solving processes. It involves understanding the capabilities and limitations of various AI technologies and making informed decisions about when to use AI for automation, augmentation, or independent agent-mediated experiences.

Subcategories

a) Goal and Task Awareness:

• Envisioning an effective goal for a project.

• Understanding the nature and requirements of the task(s) towards the defined goal.

• Ability to analyze and deconstruct a task into AI, human, and collaborative components.

• Necessary for effective integration of AI into creative workflows.

b) Platform Awareness:

• Understanding the capabilities and limitations of current AI tools.

• Knowledge of various AI platforms and their specific strengths and limitations  in relation to the project’s goal.

• Ability to evaluate AI tools based on project requirements, budget, operational and regulatory needs.

• Necessary for selecting the optimal AI tools for specific tasks.

c) Task Delegation:

• Balancing AI and human capabilities throughout a project to best realize the creative vision.

• Understanding the different affordances of each modality (Automation, Augmentation, Agency).

• Ability to assign project tasks to human and AI tools optimally.

•  Necessary for successful collaboration between human and AI in creative processes.

Description - Effectively describing a vision and/or tasks to prompt useful AI behaviors and outputs. 
Description encompasses the skills needed to effectively communicate ideas, requirements, constraints, and other aspects of creative visions to AI systems. It involves crafting clear, specific, and well-structured prompts (using a wide range of prompting techniques) and other elements that guide and enable AI tools to produce desired behaviors and outputs.

Subcategories:
a) Product Description:

• Prompting to define desired output.

• Ability to clearly articulate desired characteristics, features, and qualities of the final AI-generated output.

• Skill in translating creative vision into explicit, AI-understandable terms.

• Crucial for guiding AI tools to produce results aligned with the creator's intentions.

b) Process Description:

• Dialogic prompting to produce effective iterative collaboration.

• Ability to engage in dynamic, back-and-forth communication with AI tools.

• Skill in breaking down complex tasks into a series of smaller, manageable prompts.

• Essential for guiding AI through multi-step creative processes aligned with the human collaborator.

c) Performance Description:

• Directive prompting to define future AI behaviors and enable positive user experience.

• Ability to define how AI-generated content or systems should behave or interact with users.

• Skill in anticipating user needs and translating them into guidelines for AI behavior.

• Critical for enabling future AI-driven behaviors that are aligned with the human’s vision and values. 

Discernment - Accurately assessing the usefulness of AI outputs 
Discernment involves the critical evaluation of AI-generated outputs, understanding their quality, relevance, potential biases, and other salient characteristics. It also includes the ability to iterate and refine the collaborative process with AI tools.

Subcategories:
a) Product Discernment:

• Evaluating output quality and identifying ways to improve it.

• Ability to critically assess the quality, relevance, and effectiveness of AI-generated content.

• Skill in identifying strengths and weaknesses in AI outputs.

• Crucial for maintaining high standards in AI-assisted creative work.

b) Process Discernment:

• Assessing if the human-AI collaborative dynamic is fruitful or not and how to improve it.

• Ability to evaluate the effectiveness of the human-AI collaborative process.

• Skill in identifying which aspects of human-AI interactions are most beneficial and where improvements can be made.

• Essential for optimizing the use of AI tools in creative collaborative work.

c) Performance Discernment:

• Evaluating if AI-driven independent behaviors enable positive user experiences and how to better direct the AI to improve outcomes.

• Ability to assess the effectiveness of AI systems in independent, user-facing scenarios.

• Skill in gathering and interpreting human feedback to refine and ensure intended AI-driven behaviors and experiences.

• Essential for designing user experiences aligned with the project's vision and values. 

Diligence - Taking responsibility and vouching for final products created using AI 
Diligence refers to the responsible use of AI, including ethical considerations, transparency about AI use, and taking accountability for the final products created with AI assistance.

Subcategories:
a) Creation Diligence:

• Responsible use of AI tools, maintaining ethical and legal best practices, awareness of biases, flaws, stakeholder impacts, and other externalities

• Understanding and applying ethical principles throughout the AI-assisted creative process.

• Ability to identify and mitigate potential biases and ethical risks in AI-generated content.

• Crucial for ensuring responsible and socially conscious use of AI in creative work.

b) Transparency Diligence:

• Transparency and accountability when distributing the end product.

• Understanding of audience, industry, and legal expectations and norms around AI-generated content.

• Skill in clearly communicating the nature of AI involvement in the process.

• Essential for maintaining trust and integrity when distributing AI-assisted work.

c) Deployment Diligence:

• Taking responsibility for verifying and vouching for AI-assisted outputs, including thorough fact-checking, testing for accuracy, and validating claims.

• Implementing appropriate safety checks and testing procedures before releasing AI-assisted work.

• Understanding, managing, and assuming responsibility for potential risks and impacts of deployed AI-assisted content and/or agents.

• Essential for ensuring the quality, safety, and reliability of content and/or agents created through Human-AI interaction.