Create Rule “use-ai-responsibly/rule” (#9814)

rules/use-ai-responsibly/rule.md

@@ -0,0 +1,75 @@
+---
+type: rule
+tips: ""
+title: Do you use AI responsibly?
+seoDescription: Learn how to use AI responsibly by balancing transparency,
+  critical thinking, data privacy, and getting the best value.
+uri: use-ai-responsibly
+authors:
+  - title: ""
+  - title: Ulysses Maclaren
+    url: https://www.ssw.com.au/people/ulysses-maclaren/
+related:
+  - manage-security-risks-when-adopting-ai-solutions
+  - manage-legal-implications-of-ai
+  - mitigate-brand-risks-ai
+created: 2025-01-29T14:29:00.000Z
+guid: 3549fab7-07a1-42fe-88ad-1296be272018
+---
+Many people hesitate to rely on AI, fearing it will replace human decision-making or lead to uncritical dependence. However, the real problem isn’t using AI—it’s using it blindly. Just like any other tool, AI should be leveraged intelligently to enhance productivity and decision-making. 
+
+The best approach is to become a *centaur* — a combination of human intelligence and AI efficiency. Ignoring AI altogether is just as irresponsible as using it without critical thinking.
+
+<!--endintro-->
+
+## Be Transparent When Using AI
+
+Transparency is key when using AI-generated content. If AI is simply refining your own work, disclosure may not be necessary. However, there are cases where you should clearly communicate AI involvement:
+
+* **If significant portions of content are AI-generated** – For example, if an AI writes an entire report, readers should know that.
+* **If AI-generated content influences decision-making** – If you're making a business decision based on AI recommendations, others involved should be aware.
+* **When AI involvement impacts trust** – For instance, using AI-generated customer reviews or testimonials without disclosure is unethical.
+
+See [Do you indicate when AI helped? 🤖](https://www.ssw.com.au/rules/indicate-ai-helped/) for details.
+
+## Apply critical thinking
+
+You get out what you put in, both in context and effort. To get meaningful results:
+
+1. **Give the AI full context** – Include any important data or background information.
+2. **Give it some direction** – Include some bullet points for your initial solution to the problem.
+3. **Edit the AI generated responses** – Never assume the first response is perfect; refine it to match your needs.
+
+## Understand AI’s boundaries
+
+AI is powerful, but it has limitations:
+
+* **Fact-check information** – AI can hallucinate or present outdated info. Always verify with Google.com, Perplexity.ai, or subject matter experts.
+* **Know what AI can and can't do** – It excels at summarizing, brainstorming, and coding assistance but struggles with nuanced reasoning and real-world context. (Note: Capabilities are rapidly advancing, so keep your finger on the pulse).
+
+::: greybox
+"AI said this regulation applies, so we should proceed."
+:::
+::: bad
+Figure: Bad example - Relying on AI alone for compliance or legal matters without verifying against official sources.
+:::
+
+## Protect data privacy
+
+AI services differ in how they handle user data. Be mindful of:
+
+* **Training data policies** – Make sure you opt-out of your conversations being used to train the models.
+* **Confidentiality** – Never enter sensitive company data unless using private AI models.
+* **Secure options** – If needed, run AI models privately via Azure, a private cloud, or even offline using Ollama.
+
+## Use the right tool for the right job
+
+Maximize AI’s potential by using the right tools for the right tasks:
+
+* **Use ChatGPT, Gemini, Claude, etc.** – For emails, web content, marketing materials, documentation, and training resources.
+* **Use AI voice mode** – It’s great as a sounding board or to explain concepts.
+* **Use GitHub Copilot or Cursor** – For generating boilerplate code, auto-completions, unit tests, and debugging.
+
+## Be proud of your AI use
+
+AI isn’t something to hide or fear. When used correctly, it enhances efficiency and decision-making. The key is to use it *responsibly* — with transparency, critical thinking, and respect for privacy.