This is a basketball blog where people come to get their Phoenix Suns news, share opinions, and discuss what’s happening around the team. Maybe you don’t take it too seriously. I do. I care about this place, and I care about the conversations we ignite within the Bright Side of the Sun community.
Normally, I’m talking about roster additions, cap sheet strategies, projections, regressions, and theories about the Phoenix Suns. Today, I’m coming to you with something closer to a public service announcement.
I’m going to try to make you a better fan. That might sound arrogant in and of itself, but this is something that needs to be said.
Stop using AI outputs on statistical data as fact without researching the statistics yourself. It’s lazy. More importantly, it’s often incorrect.
We’re living in an interesting time when it comes to access to information. As a 43-year-old, I’ve grown up through the information age and had the benefit of experiencing the analog era before the internet, when you had to earn your information. If you wanted to know what the Phoenix Suns’ record was in 1976, there wasn’t a search bar waiting to hand you the answer.
The library was your best friend, and within the confines of that library, you had to understand how the system worked in order to find what you were looking for. I doubt many people under the age of 35 even know what the Dewey Decimal System is, but essentially, it was the pathway to finding information.
Then the internet arrived, and we entered that strange period between the invention of the internet and the rise of the smartphone. That happened to coincide with my teenage years and early 20s. I still remember sitting at bars having debates about players, where they went to college, what their impact was on the game, and how they affected a team. We did it without the benefit of pulling out a smartphone to verify everything in real time. And sure, those debates could get heated because there wasn’t always a final answer available in the moment. You couldn’t immediately confirm that Devin Booker shot 33% from beyond the arc last season.
Then smartphones became part of everyday life, and those debates changed. Now we had access to data, and we could use that information to support an argument. It was no longer only about who was right or who was wrong. It became about why they were right or why they were wrong. But in the same breath, you still had to know where and how to look for the correct information. Sites like Basketball-Reference and the statistical filters at NBA.com became the source of truth.
And now we’ve entered an era where artificial intelligence provides what feels like unlimited access to information, and people often assume whatever it gives them is correct.
In the past, you couldn’t Google a detailed statistical question and expect a clean answer. If you searched for something like the offensive rebounding percentage of the 2019-20 Phoenix Suns (the answer is 33.1%, by the way), you’d probably get a list of websites that might contain the information. You still had to find it, verify it, and understand what you were looking at.
Now, LLMs spit back answers like you’re having a conversation with a friend. And because the response sounds confident and certain, people tend to accept it as fact. In this day and age, when we have access to in-depth information on sites like B-Ball Index, Cleaning the Glass, and Second Spectrum, we have people citing ChatGPT’s half-assed, incorrect output as factual information.
That’s where my public service announcement comes in, especially for the Bright Side community that has faithfully followed this site for years, commented on posts, and exchanged thoughts, ideas, and statistics with one another. Too often lately, I’m seeing someone try to make a point in the comments by citing statistics generated by artificial intelligence. And they’re unbelievably wrong.
I felt this firsthand not too long ago when I was online after putting together an article comparing how Devin Booker and Jalen Brunson have performed in the clutch throughout their careers. I posted my findings after actually mining the data. I went toNBA.com, pulled up the clutch statistics, and went through Booker’s numbers year by year. I then put everything into a spreadsheet so I had the totality of the data in front of me and could make the comparison accurately.
I presented those findings, and this was the response I received on Twitter.
When I asked where the information came from, the answer was an AI tool. Of course.
That’s really what I’m here to provide you with today. Some knowledge. Don’t rely on ChatGPT, Gemini, Claude, or any other AI tool to give you detailed statistical answers and automatically assume they are correct. Too often, those responses are pulled from secondary sources, community discussions (like Reddit), or incomplete data. That is not the same thing as actually mining the statistics yourself.
They also struggle when the question requires more complex statistical work. In the example above, it took me about 45 minutes to go through each season, copy the data, paste it into a spreadsheet, format it, and identify what I was actually looking for: how Devin Booker and Jalen Brunson have performed in the clutch throughout their careers.
AI can absolutely help once you give it the data and tell it to start crunching numbers. That part is useful. That is a tool doing what a tool is supposed to do. The problem is that most people are not doing that. They are pulling a phone out of their pocket, typing a simple question about Devin Booker’s clutch statistics, watching an answer materialize on the screen in 3 seconds, and accepting it as fact. No digging, no checking, no wondering where the number came from. The machine said it, so into the bloodstream it goes.
That is where the frustration comes in for me. I take the time to find the right answers. I hunt down the data, organize it, verify it, and try to give you information that is actually correct. So when all of that gets dismissed because an AI tool spat out something different, yeah, that gets frustrating. And quite frankly, it gets a little scary.
There is an entire generation growing up with this kind of access to information. Answers arrive instantly, polished and confident, whether they are right or completely detached from reality. Misinformation no longer has to look suspicious. It can wear a nice suit, speak in complete sentences, and hand you a statistic with enough confidence that you never think to question it. That is the danger.
At some point, you have to learn how to find the information yourself. You have to understand where it came from. You have to earn the information. That is what we should all be trying to do. Understand what is correct. Understand what is not. Most importantly, learn how to tell the difference.
AI isn’t going away, nor should it. It can be incredibly useful when you understand what it does well and where it can fail. But convenience should never replace verification. If you’re going to use statistics to support an argument, know where those numbers came from. Check the source. Understand the methodology. Do the work when necessary. Because access to more information only makes us smarter if we remain willing to question it.
There’s another part of this whole situation that scares me. When you go to places like Google to look up information, you’re no longer always being presented with links that lead you toward detailed explanations of what you’re searching for. Instead, you’re often being given an AI summary, in Google’s case, through Gemini.
Take this article, for example. My assumption is that if you go to Google and search “Phoenix Suns ChatGPT”, there’s a chance you’ll get an AI overview that links back to this article. The problem is that too often people won’t take the time to click on that link. They’ll read the two-sentence summary, take it at face value, and move on. And in doing so, they can completely miss the nuance that exists within actual reporting, an article, a thought, an idea, or a blog post.
That’s the part that scares me on a personal level because this is what I love to do. I love writing about the Phoenix Suns. I love taking these thoughts and ideas and presenting them to the world. And more and more, it feels like those thoughts and ideas are being compressed, summarized, and muted by AI before people ever reach the source.
That may ultimately be the larger cost of convenience. Information becomes easier to access while the people who actually create it become easier to overlook. Statistics lose their context. Arguments lose their nuance. Writing becomes a collection of bullet points generated before the reader ever reaches the person who did the work. AI can help us find, organize, and understand information, but it shouldn’t become a substitute for curiosity.
Click the link. Read the article. Check the data. Find the source. Earn the information. And for the love of God, don’t be an idiot and use AI output as fact. It really ticks my bombs.
