The San Mateo Daily Journal does not use artificial intelligence in its writing or reporting, neither will it in the immediate future. It also doesn’t accept any submissions that are written with the assistance of AI.
I had assumed, wrongly, that people would know not to submit guest perspectives or letters to the editor without AI but that is not the case. To add to my growing list of tasks, I now must run submissions through several AI detectors to make sure it did not assist the letter writer. As a writer, I would be ashamed to use it. Others, however, are not writers and likely are more skilled than me in other aspects of life. If one struggles with writing, the appeal of AI is real. It’s a way to make your thoughts presentable and succinct. However, it is my opinion that it is not authentic and sets a poor model for others. If we now outsource our thought processes to a computer, we are abdicating the very thing that makes us human. It is cheating, in a way.
I know others may not see it that way, and that’s OK. But for our purposes, here at the Daily Journal, we believe those who have thoughts should put pen to paper, or fingers to keyboard, as process those thoughts themselves.
Submitting a set of thoughts to AI and asking it to come up with an essay from it may come across as nice and easy, but there is something about the writing process that forces the writer to fully think through their ideas on their own. In writing, an idea could arise, or there could be a completely different point of view that comes through in arguing for a particular point of view. Feeding an idea into a machine will not do that. It is limited in scope.
There are many reasons for someone to think about submitting a letter to the editor or a guest perspective. More often than not, it is because they have a particular point of view they seek to convey and to convince others to think the same way while also spurring them into action. If you haven’t mastered the “why” to your piece, it’s likely you won’t spur anyone to anything but you might cause those who think similar to continue to think that way. In short, you’ve done nothing.
So the idea of not using AI in Daily Journal submissions is really rooted in the concept that it is important to go through the writing process yourself as a way to fully flesh out the ideas that come from you. As with anything, it’s OK to bounce ideas off others as that’s a way to get new ideas or figure out new thoughts.
Recommended for you
This is the essential concept behind why AI-assisted writing is not allowed in academic settings. Writing itself is an extension of a thought process, and building and growing ideas through that process is essential to learning — and that should not be just reserved for school, but rather throughout a lifetime.
We are also available to you, our readers, to help with your work as well. Within reason, of course. But you can also look toward your own community for feedback. Friends, colleagues and family know you and can help build your ideas. This is a very human thing to do, and builds immediate community connections as well.
Some news agencies are also using AI in their reporting. We believe it’s OK for them, as long as it is labeled as such. The challenges for news agencies are myriad right now, and some have to do what they have to do to survive. However, it appears to us at least to be a poor substitute for actual reporting in which the reporter watches and contextualizes civic action rather than just regurgitating minutes in bullet form. As far as understanding specific actions from a government meeting, sure, but putting it into larger context and humanizing the language is the job of a reporter. Also, there are times a discussion could yield a further story. As an example I give you a priority discussion from the San Bruno City Council in which uneven parking enforcement was addressed. While that might be a quick line item in a larger AI meeting minutes report, it also merited further investigation by our reporter.
Besides, we are still humans. We bring that element to our reporting. It’s not simply conveying information, it’s delivering it in a way to makes sense, that provides meaning and advances community conversations.
So AI is not for us here at the Daily Journal. At least not yet, because it doesn’t make sense for our overall purpose here, which is not just sharing information but making sense of it. AI is a pretty good tool overall, just not for us here in our newsroom.
Thanks to the San Mateo Daily Journal for always being a place where ideas can grow, debate can flourish, and individuals, with their own non-AI-generated thoughts, can engage to make our community stronger!
Thanks for your column today, Mr. Mays, relating the SMDJ’s position on AI usage. It is an admirable position and we can only hope that contributors follow the “rules.” As for checking contributions to the DJ, perhaps you can modify guidelines so contributors do their homework and include screenshots or evidence that their submission has been run through AI detectors and meet a certain standard. Now if only we could program an AI to detect fake news and lies – perhaps that is the next evolution of AI. Of course, it means many SMDJ LTE’s and guest perspectives will never see the light of day. But that wouldn’t be so bad…
AI detectors don't actually "detect" AI. Instead, they use statistics to calculate how much a text resembles an AI model's output. AI detectors guess the next word in your sentence and gives it a score on how well it predicted it correctly (perplexity). Then it looks for burstiness and measures the variation. Humans write in bursts...long, complex sentences followed by short ones. I thought it used Bayesian processes, but AI helped me better understand my inaccurate assumption.
I must thank you for irony though. A human editor abdicating his own "authentic" judgment to AI detector in order to prove that readers didn't abdicate their own judgment to an AI tool. :)
My sixth‑grade daughter was recently flagged by her English teacher for “using AI” on a writing assignment. The only problem? She has no access to the internet or any AI tools. The accusation was dead wrong.
As others have pointed out, AI detectors are far from foolproof. I imagine SMDJ editor Mays was once wary of Wikipedia, or even the internet itself. To those of us who write reasonably well, AI can be a helpful editing tool. For those who struggle, it can be transformative. You have to be looking through a very narrow aperture to miss the benefit of having language expression skills suddenly become more universal.
AI is a tool, no different in spirit from Spell Check or a thesaurus. My physics teacher once made us use slide rules so we wouldn’t “become dependent on calculators". Editor Mays might appreciate that kind of anachronistic thinking.
I appreciate this policy for the stories in the Journal, and I appreciate the effort you put into writing an explanation of it. I don’t think commenters should use AI but trying to detect it might not be worth the effort.
As someone who's worked as a writer and editor for 35 years, I agree completely with the Journal's position on AI. Using AI can be helpful in researching a topic and in editing the final copy, but it's got to be a human who puts the ideas down on the page, if it's a human who's going to sign their name to the letter or article. In response to the commenter who's thankful for the help that AI can give to less confident writers: I agree! In this capacity, AI still needs to be kept to the role of editor, not author. Even the least fluent writer needs to lay out their basic argument using whatever words (in whatever language) they can find in their own head. They can then run their rough text through an AI tool like ChatGPT to improve their writing, and even ask the AI to poke holes in their logic, but ultimately the author needs to take responsibility for every word of their final draft.
Keep the discussion civilized. Absolutely NO
personal attacks or insults directed toward writers, nor others who
make comments. Keep it clean. Please avoid obscene, vulgar, lewd,
racist or sexually-oriented language. Don't threaten. Threats of harming another
person will not be tolerated. Be truthful. Don't knowingly lie about anyone
or anything. Be proactive. Use the 'Report' link on
each comment to let us know of abusive posts. PLEASE TURN OFF YOUR CAPS LOCK. Anyone violating these rules will be issued a
warning. After the warning, comment privileges can be
revoked.
Please purchase a Premium Subscription to continue reading.
To continue, please log in, or sign up for a new account.
We offer one free story view per month. If you register for an account, you will get two additional story views. After those three total views, we ask that you support us with a subscription.
A subscription to our digital content is so much more than just access to our valuable content. It means you’re helping to support a local community institution that has, from its very start, supported the betterment of our society. Thank you very much!
(6) comments
Thanks to the San Mateo Daily Journal for always being a place where ideas can grow, debate can flourish, and individuals, with their own non-AI-generated thoughts, can engage to make our community stronger!
Thanks for your column today, Mr. Mays, relating the SMDJ’s position on AI usage. It is an admirable position and we can only hope that contributors follow the “rules.” As for checking contributions to the DJ, perhaps you can modify guidelines so contributors do their homework and include screenshots or evidence that their submission has been run through AI detectors and meet a certain standard. Now if only we could program an AI to detect fake news and lies – perhaps that is the next evolution of AI. Of course, it means many SMDJ LTE’s and guest perspectives will never see the light of day. But that wouldn’t be so bad…
AI detectors don't actually "detect" AI. Instead, they use statistics to calculate how much a text resembles an AI model's output. AI detectors guess the next word in your sentence and gives it a score on how well it predicted it correctly (perplexity). Then it looks for burstiness and measures the variation. Humans write in bursts...long, complex sentences followed by short ones. I thought it used Bayesian processes, but AI helped me better understand my inaccurate assumption.
I must thank you for irony though. A human editor abdicating his own "authentic" judgment to AI detector in order to prove that readers didn't abdicate their own judgment to an AI tool. :)
My sixth‑grade daughter was recently flagged by her English teacher for “using AI” on a writing assignment. The only problem? She has no access to the internet or any AI tools. The accusation was dead wrong.
As others have pointed out, AI detectors are far from foolproof. I imagine SMDJ editor Mays was once wary of Wikipedia, or even the internet itself. To those of us who write reasonably well, AI can be a helpful editing tool. For those who struggle, it can be transformative. You have to be looking through a very narrow aperture to miss the benefit of having language expression skills suddenly become more universal.
AI is a tool, no different in spirit from Spell Check or a thesaurus. My physics teacher once made us use slide rules so we wouldn’t “become dependent on calculators". Editor Mays might appreciate that kind of anachronistic thinking.
I appreciate this policy for the stories in the Journal, and I appreciate the effort you put into writing an explanation of it. I don’t think commenters should use AI but trying to detect it might not be worth the effort.
As someone who's worked as a writer and editor for 35 years, I agree completely with the Journal's position on AI. Using AI can be helpful in researching a topic and in editing the final copy, but it's got to be a human who puts the ideas down on the page, if it's a human who's going to sign their name to the letter or article. In response to the commenter who's thankful for the help that AI can give to less confident writers: I agree! In this capacity, AI still needs to be kept to the role of editor, not author. Even the least fluent writer needs to lay out their basic argument using whatever words (in whatever language) they can find in their own head. They can then run their rough text through an AI tool like ChatGPT to improve their writing, and even ask the AI to poke holes in their logic, but ultimately the author needs to take responsibility for every word of their final draft.
Welcome to the discussion.
Log In
Keep the discussion civilized. Absolutely NO personal attacks or insults directed toward writers, nor others who make comments.
Keep it clean. Please avoid obscene, vulgar, lewd, racist or sexually-oriented language.
Don't threaten. Threats of harming another person will not be tolerated.
Be truthful. Don't knowingly lie about anyone or anything.
Be proactive. Use the 'Report' link on each comment to let us know of abusive posts.
PLEASE TURN OFF YOUR CAPS LOCK.
Anyone violating these rules will be issued a warning. After the warning, comment privileges can be revoked.