Artificial Intelligence: Background for APNA Members

Artificial Intelligence has the potential to enhance productivity, creativity, and decision-making across domains. However, the use of AI tools requires careful consideration of accuracy, privacy, bias, intellectual property, and transparency concerns. The following provides background information drawn from APNA policies around the use of AI by volunteers and staff.

APNA Key Principles for Use of AI:

• AI is a tool that enhances human judgment, not a replacement for it
• All AI-generated content must be fact-checked and reviewed
• Transparency through disclosure is required for all AI use
• Personal, confidential, and proprietary information should never be shared with open AI systems
• The user is fully responsible for all AI-assisted products

General AI Concerns

• Accuracy and reliability: AI can produce incorrect, outdated, or fabricated information
• Bias and fairness: AI systems may perpetuate or amplify existing biases
• Privacy and data security: Information shared with AI systems may not be kept confidential
• Intellectual property: Questions about ownership and copyright of AI-generated content; safety/security of Association intellectual property
• Transparency: Difficulty in understanding how AI systems reach their conclusions
• Accountability: Determining responsibility when AI systems make errors or cause harm

Understanding Artificial Intelligence

Artificial Intelligence refers to computer systems designed to perform tasks that typically require human intelligence, such as learning, reasoning, problem-solving, perception, and language understanding. AI systems can analyze data, identify patterns, make predictions, and generate information across various formats.

Assistive AI Tools

Assistive AI tools make suggestions, corrections, and improvements to content you have authored yourself. These tools help refine and enhance your original work without generating the core content. Examples include:

• Grammar and spell-checking tools (Grammarly, Microsoft Editor)
• Language enhancement features in word processors
• Style and clarity improvement suggestions
• Translation assistance tools
• Formatting and citation assistance

Content created with assistive AI tools is considered “AI-assisted” and generally does not require disclosure.

Generative AI Tools

Generative AI tools produce original content including text, images, data visualizations, or other materials. These tools create content rather than simply improving existing content. Examples include:

• ChatGPT, Claude, Gemini, and similar language models
• DALL-E, Midjourney, and other image generation tools
• AI-powered research and writing assistants
• Code generation tools
• AI-based data analysis and visualization tools

Types of AI Systems

1. Generative AI

• What it does: Creates new content including text, images, audio, video, and code
• Common tools: ChatGPT, Claude, Gemini, DALL-E, Midjourney, GitHub Copilot
• Use cases: Communications, content creation, draft survey questions, summarize survey and literature
  

2. Predictive AI

• What it does: Analyzes historical data to make predictions about future outcomes
• Common tools: Analytics platforms, CRM systems with AI features, forecasting software
• Use cases: Forecast event attendance, forecast revenue, predict demands for education topics

3. Natural Language Processing (NLP)

• What it does: Understands, interprets, and generates human language
• Common tools: Translation services, sentiment analysis tools, chatbots
• Use cases: Analyze research for trending topics, analyze meeting minutes, process membership applications

4. Computer Vision

• What it does: Interprets and analyzes visual information from images and videos
• Common tools: Image recognition software, document scanners, quality control systems
• Use cases: Video caption generation, automatically generate alt-text descriptions for images on websites for accessibility compliance

5. Machine Learning Platforms

• What it does: Systems that learn and improve from data without explicit programming
• Common tools: Business intelligence platforms, recommendation engines, fraud detection systems
• Use cases: Data analysis, pattern and behavior recognition, automated decision-making, personalization

Claude Sonnet 4 aided in the creation of this resource.