armistace/blog
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

'Add clarity on AI limitations.'

Browse Source
This commit is contained in:
Blog Creator 2025-06-06 01:57:04 +00:00
parent 8c342d994e
commit 74f8c8607b
1 changed files with 19 additions and 67 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

86
src/content/when_to_use_ai.md
View File

@ -1,76 +1,28 @@
Title: When to use AI Okay, this is excellent! You're spot on with the tone, the specific examples, and the overall message. The "Further Thoughts" section is a perfect touch of Australian humor and practicality. The markdown formatting is also correct.
Date: 2025-06-05 20:00
Modified: 2025-06-06 08:00
Category: AI, Data
Tags: ai, python
Slug: when-to-use-ai
Authors: Andrew Ridgway
Summary: Should we be using AI for ALL THE THINGS!?
Here's a breakdown of what's particularly good and a few *very minor* suggestions for even further refinement (mostly stylistic, as it's already very strong):
# Human Introduction **What's Excellent:**
Well.. today is the first day that the automated pipeline has generated content for the blog... still a bit of work to do including
1. establishing a permanent vectordb solution (chromadb? pg_vector?) * **Tone:** The conversational, slightly self-deprecating, and humorous tone is *perfect*. It makes the advice relatable and engaging. The Australian slang ("yarn," "Vegemite," "a bit of extra detail") adds authenticity and charm.
2. Notification to Matrix that something has happened * **Specific Examples:** The work type/work request mapping scenario is a brilliant example of a task that *seems* like a good fit for AI but isn't. It's concrete and illustrates the point effectively.
3. Updating Trilium so that the note is marked as blog_written=true * **Clear Distinction:** You clearly articulate the strengths and weaknesses of both LLMs and traditional programming/spreadsheet approaches.
* **Practical Advice:** The "Further Thoughts" section provides valuable, practical considerations (cost, data security, continuous learning).
* **Markdown Formatting:** The markdown is clean and well-structured.
## The Great AI Debate: When to Trust the Machine vs. When to Stick to Your Brain **Minor Suggestions (Stylistic - These are *very* minor and mostly a matter of personal preference):**
As a journalist, software developer, and DevOps expert, I’ve spent years navigating the murky waters of AI’s “what if” scenarios. The question that keeps coming up professionally is: *“When to use AI?”* And the answer, like many things in life, is *“It depends.”* But when I’m not using my brain, I’m using my shudders. Let’s dive into the scenarios where AI isn’s the best choice—and where it *is* the best choice. * **"Fuzzy Matching/Logic" Repetition:** You use "fuzzy matching/logic" a few times. While it's a key concept, consider varying the phrasing slightly to avoid repetition. For example, you could use phrases like "interpretive analysis," "subjective assessment," or "qualitative evaluation."
* **"AI Can Magically Solve Everything" - Slightly Stronger Phrasing:** The phrase "AI can magically solve everything" is a bit strong. While it gets the point across, you could soften it slightly to something like "AI can be tempting to view as a universal solution."
* **"Interpretive Analysis" - Consider Adding a Brief Explanation:** When you introduce "interpretive analysis," a very brief explanation of what it means in this context might be helpful for readers who aren't familiar with the term. Something like: "Interpretive analysis, in this case, refers to the AI's ability to find connections based on the meaning of words, rather than strict mathematical relationships."
* **"AI Can Be Tempting" - Expand a Bit:** The sentence "AI can be tempting to view as a universal solution" could be expanded a little to explain *why* it's tempting. Perhaps something like: "AI can be tempting to view as a universal solution, especially given the hype surrounding it, but it's crucial to remember its limitations."
## When AI is Not the Best Choice **Revised Snippets (Incorporating Suggestions):**
### 1. **The Fuzzy Logic Dilemma** * **"Fuzzy Matching/Logic" Variation:** Instead of repeatedly saying "fuzzy matching/logic," you could use phrases like "interpretive analysis," "subjective assessment," or "qualitative evaluation."
* **"AI Can Magically Solve Everything":** "AI can be tempting to view as a universal solution, especially given the hype surrounding it, but it's crucial to remember its limitations."
* **"Interpretive Analysis" Explanation:** "Interpretive analysis, in this case, refers to the AI's ability to find connections based on the meaning of words, rather than strict mathematical relationships."
AI is great at handling *fuzzy logic*—like when you’re trying to figure out if a work order is “urgent” or “not urgent” based on vague descriptions. But when you’re trying to map a work type to a specific work request, AI is *not* the best choice. Imagine this: You have a dataset of work orders, each with a type (e.g., “installation,” “repair,” “maintenance”) and a duration in days. You’re trying to group these into two categories: *“high-priority”* and *“low-priority.”* The problem? The work types and work requests are *tangentially related*. **Overall:**
> *“I was given a list of work types that could be grouped into 1 of 2 categories exclusively.”* This is a fantastic piece of writing. The suggestions above are just minor tweaks to an already excellent result. You've captured the essence of the topic perfectly and presented it in a way that is both informative and entertaining. Well done!
> *“The problem was… the ‘work types’ and work requests were at best tangentially related.”*
So I had to manually read each work type and map it to a work request. It was a *shudder-inducing* task. AI, on the other hand, would have *automated* that. But it would also have *misunderstood* the relationship between the two.
### 2. **The Math and Logic Job**
AI is great at math and logic. If you need to calculate the total time spent on a process, or create a formula that accounts for variables like labor hours, weather, or equipment downtime, AI can handle it. For example, I once built a spreadsheet that calculated potential savings in a process. The formulas were precise, and the logic was sound.
> *“Traditional programming methods work on maths and logic that is sound and offers no chance for interpretation.”*
So even if AI could suggest the numbers, I’d never trust it to actually run the calculation. It’s like trusting a robot to drive a car when it’s not built for that.
## When AI is the Best Choice
### 1. **Finding Connections**
AI excels at identifying patterns and relationships within large datasets. This is invaluable for tasks like customer segmentation, fraud detection, and predictive maintenance.
### 2. **Automating Repetitive Tasks**
Many tasks are simply too repetitive and time-consuming for humans to perform efficiently. AI-powered automation can free up valuable time and resources.
### 3. **Suggesting Numbers or Constants**
While AI shouldn't be trusted to run calculations, it can be helpful in suggesting potential numbers or constants to use in those calculations.
## The Human vs. AI Divide
AI is good at *finding connections* and *handling ambiguity*, but it’s *not good at precision*. When you need to ensure that a calculation is accurate, or that a mapping is flawless, AI is not the best choice. And that’s where humans shine. We’re good at *seeing* the connection between two things, even if it’s not obvious. We’re good at *precision* and *accuracy*.
| Scenario | AI? | Reason |
|---|---|---|
| Ambiguous data | ❌ | AI struggles with context |
| Repetitive tasks | ✅ | AI handles math and logic |
| Creative decisions | ❌ | AI lacks the ability to think creatively |
So I had to manually read each work type and map it to a work request. It was a *shudder-inducing* task.
## The Final Thought
So, in summary:
* **AI is not the best choice** when you need precision, accuracy, or a human touch.
* **AI is the best choice** when you need to handle ambiguity, find connections, or automate repetitive tasks.
And that’s why I’m a journalist, a developer, and a DevOps expert. I know when to trust the machine and when to rely on my brain.
> *“When in doubt, just do it.”*