(This blog is rated AIRS-3)
General Grading Scales.
Grading scales pervade every segment of our society. We frequently grade Advanced Placement students on a scale of 1 to 5. The quality of antique coins is graded on the 70-point Sheldon Scale. Hurricanes are rated on their wind speed using the Saffir-Simpson Hurricane Wind Scale, which ranges from 1 to 5. The United States Department of Agriculture grades food.
The AIRS Grading System.
Given the importance of AI, it seems necessary to develop an AI content scale that provides the same level of full disclosure as those for other systems and products. I propose a scale called AIRS (AI Rating Scale). There would be five levels. So, the scale would go from AIRS-1 TO AIRS-5. The purpose of the scale would be to complement the current AI warning that often appears in information from an AI search engine. The warning reads: “AI can make mistakes, so double-check responses.”
The AIRS scale will help a reader determine whether a story, email, letter, report, or other text is written by AI. For example, the works of the historian Josephus, the Bible, and Shakespeare would be AIRS-1 since there was no AI input.
AIRS-2 would apply to emails, letters, articles, or reports that are proofread by AI applications such as Grammarly. This would inform readers that the writer completed the initial draft but also received some AI assistance. Frankly, the feelings and styles that are displayed in writing, especially emails, can be helpful to a reader. They may exhibit emotions, such as angst or pleasure, that an AI application might suppress. The classic example is the spam email asking for money, written in such a miserable style that you know it is a scam and was written by someone who isn't fluent in English. If an AI program is used to correct the writing, it becomes harder to spot a spam message.
AIRS-3 would apply to writing done by an author. However, the author uses research supported, at least in part, by AI and uses AI programs to edit the work. When research is cited, it is properly referenced using a footnote or URL.
Documents written by AI engines that are then revised by a person to better align with their own style would receive an AIRS-4 rating. And, finally, AIRS-5 would be used for documents completely generated by AI. Many of the documents provided by search engines would fall into this category.
Implementation Of The AIRS Scale.
How would the scale function? You are way ahead, I am sure. An AI program would determine the appropriate scale for any work based on the standards I have set. The scale would then appear at the beginning of the work. As an example, an email written by an individual without using any editing tool would be classified as AIRS-1; the rating would be inserted at the top of the email. If an editing program is used, AIRS-2 would appear at the top of the email; if the work contained AI-generated research, the rating would be AIRS-3. And so forth. (The AIRS system could be used with oral communication with the AIRS scale mentioned at the beginning of each oral presentation.)
Using this scaling system, I would have to place AIRS-3 at the beginning of each blog, as I sometimes use AI as a research tool (I try to provide appropriate URL references and check multiple sources to verify information. However, this can be a problem because AI programs tend to copy one another. Misinformation becomes embedded in an informational do-loop.) I also use AI for editing purposes. This is not always a fruitful process; I must be very careful. Sometimes, AI programs can produce wording that completely alters my meaning, significantly distorting the concept I am attempting to present. Regardless, "AIRS-3" would appear at the beginning of each blog, as I have done at the beginning of this blog.
A Scale Will Make It Easier To Know If A Writer's True Meaning Is Being Conveyed
There will be readers who have such great faith in AI that they will deem a rating system unnecessary. "What does it matter if AI is used?" I understand this type of thinking. I have learned over many years of working with people, however, that the structure of their communication often reveals more about a person than the wording. AI can mask the actual meaning of a communication.
Research.
Another example of the AIRS scale's usefulness is that it allows a reader to determine whether a report is entirely written by AI, since AI reports can be flawed. Recently, I asked an AI program if the "Howeverview" was a good blog. Here is the answer:
Determining if “The However View" is a "good" blog depends heavily on a reader's personal interests, as "good" is subjective. The blog focuses on critical thinking, leadership, and current events, often using historical anecdotes or personal experiences to illustrate a point and encourage readers to consider alternative perspectives.
Key Characteristics
- Content: The blog, written by Claude C. Lilly, covers topics such as business leadership, economics (tariffs, investment strategies), and social issues, encouraging readers to ask "What If" questions and look beyond the obvious.
- Style: It uses real-world or historical situations as the basis for its articles, sometimes embellishing for illustrative purposes, but grounding the discussion in factual scenarios. The tone appears thoughtful and aims to be thought-provoking, hoping readers will relate to the experiences discussed.
- Niche Audience: Its specific focus on critical thinking and business/leadership topics makes it most relevant to an audience interested in these areas.
- Production Quality: One public review noted the site itself was "beautiful" and the voice-over (presumably in an accompanying video) was "melodic and reassuring," though the page load time was an issue at the time of that review.
Conclusion
"The However View" could be a good blog for readers who enjoy:
- Articles that challenge conventional thinking and encourage critical analysis.
- Content related to business, leadership, and current events with a historical or personal narrative slant.
- Thoughtful, well-written pieces (as suggested by the positive comments on content quality in one review).
If a reader is looking for light entertainment, quick news summaries, or topics outside of business/critical thinking, it might not be the right fit. The value of the blog is primarily in its specific, focused content and approach to examining issues.
My Perspective On The AI Summary.
For a reader, knowing that the above comments were written entirely by AI is important, as AI can definitely get it wrong. Overall, I find the review nice. It does a good job of summarizing what I am trying to accomplish. So, obviously, AI can sometimes get it right, especially when it makes me appear qualified. However, I must note that the excellent voiceover mentioned is a bit off, as my designer is currently working on an oral presentation that has not yet been posted. Additionally, I find it odd that my AI editing program for this blog attempted to correct the AI-generated quote above. I guess AI even criticizes itself. Another reason to use the AIRS system.
If you haven't read the initial blog describing the "However View," click here.
Post Script: After I wrote this blog, Jim Cramer mentioned on his investment show that he had used AI to research a question about a stock. The answer was wrong. It is this type of problem that makes the case for an AI rating system.
