Wednesday, March 29, 2023

AI coming to Trackside Treasure

****BREAKING - see bottom of post for an iportant update****
Although I have enjoyed blogging for nearly fifteen years, it is A LOT of work. Just a TON of work! There's relentless research, reams of writing and finicky photography. Unless you've done it, you don't realize how much personal energy it requires. I'm trackside tuckered. Fortunately in blogging, as in many other fields, the technology is allowing us to do more with less. Whether it's self-checkouts at the store, robots in manufacturing or voice-recognition software, work is supposed to get easier, not more difficult.

With the advent of AI*, we all have an ability to harness the awesome power of technology to make life easier and simpler. It's been a tough decision I've been wrestling with for some time, but I'm going to make greater use of AI here on Trackside Treasure. For every three AI posts, I will write one the traditional way. The good news is that most Trackside Treasure readers will notice absolutely no difference. 

I've been using Chat GPT, an emerging and still-evolving technological tool. I found that GPT refers to Generative Pre-Train Transformer. Perfect - just like the train transformer on my layout. Perhaps. Anyway, I jumped in with both feet and I'm pleased to present the following foreshadowing of Trackside Treasure's transformation. I've gone from a labour-intensive, human-centric platform to a sleek, high-tech and readable new blog format almost instantly. Thanks for being along for this transformational trainsition. Enjoy the future by checking out these sample passages below. First, Chat GPT tells us all about VIA Rail Canada:
Then, Chat GPT tells all about Trackside Treasure!
Here's Chat GPT's take on model railroads?
Remember, these are early iterations. I expect the technology to become more honed over the next few weeks. Can you imaging scrolling down through each Trackside Treasure post and revelling in paragraph after paragraph of the new-age blogging that this represents? And it's so incredibly easy for me. I just ask a couple of quick questions and the technology spits out entire passages in just a few seconds! Wow! What about GP9's and Rapido Trains?

I think you'd have to agree that these screenshots give just a small glimpse into the future of this blog. I'm excited by the change and I hope you will be, too! I only wish I had discovered this years ago. Think of all the ink and crumpled up pieces of paper I could have saved, not to mention all the pencil erasers I chewed off while thinking what to write next. That is time I'll never get back. And erasers.

But just as you can't un-ring a bell, you can't over-think each and every blog post. Chat GPT sure doesn't. It's just wham, bam, text slam! Thank you AI - *Aprilficial Intelligence!

**** BREAKING update! April 1 at noon! ****

Trackside Treasure is pumping the brakes on the AI switchover. More than a thousand notable signatories, including Twitter, Tesla, and SpaceX boss Elon Musk, and Steve Wozniak, co-founder of Apple, have signed an open letter calling on all artificial intelligence (AI) labs to pause the training of systems more powerful than GPT-4.

The letter states there isn’t enough planning and management of AI advances. The opposite is supposedly happening – AI labs have been locked in an “out-of-control race to develop and deploy ever more powerful digital minds that no one – not even their creators – can understand, predict, or reliably control.” Even though the signatories say they’re not calling for a complete pause on AI development in general, they add that experts should use the letter’s proposed half-year hiatus to develop and implement shared safety protocols.

It sounds like Trackside Treasure was heading down the wrong track. I can barely control my own intelligence, and the Aprilficial intelligence could go out-of-control and start posting only CP posts, or posts on O-scale modelling, or only posts on Athearn blue box kits of yore. I'll be reassessing the switch to AI next April.

Running extra...

I finished reading Maggie Haberman's excellent and comprehensive book Confidence Man (above, with CNN's Kaitlan Collins).

"The man who made racist comments and then insisted people had misunderstood him. He was interested in money, dominance, power, bullying and himself. He treated rules and regulations as unnecessary obstacles rather than constraints on his behavior. He lost his temper suddenly and abusively, directing his ire at one aide in a roomful of others. He sought an endless stream of praise. He created an enviroment perpetually beset by rivalries. He encouraged people to take risky actions in his name, and demanded they prove hthemselves to him over and over." 

This was not a mob boss nor a dictator, but rather a former US president who acted like both. As Haberman puts it, "He became the subject of more books during and immediately after his presidency than...any other one-term president...explored his moods, his dysfunctional management style, and how he reached decisions on critical policy matters.

He has only had a handful of moves throughout his entire adult life. There is the counterattack, the quick lie, the shift of blame, the distraction or misdirection, the outburst of rage, the performative anger, the indecisiveness masked by a compensatory lunge, the backbiting about one adviser with another. The challenge is figuring out at any given moment which trick he is using.

No-one in U.S. history has won the White House without first holding high public office. Grant and Eisenhower had never been elected to anything, but were Generals of the Army during wartime; all the rest had been state or national elected figures before going on to the White House.

I have to wonder how a great nation was so hoodwinked by someone so morally bankrupt yet completely electible; who once in office shredded norms while wrapping himself in a flag that represented a republic built on every tenet he thoroughly trashed. Coming soon to Court TV.


Michael said...

I've learned never to trust a post around April 1!

Eric said...

Chat GPT is handling the comment section this post.

Thank you for your comment. Ottawa is a hotbed of French and English. Some commenters comment in French saying "comment allez-vous". This poast is keeping a comment tally view. That is the value of AI generated comment handling. It is accurate, more logical and of great use to readers in many languages. As we say in French, O Review!

Mike Galt said...

I like the old way of writing with thousands of monkeys and typewriters, not this new fangled crap method that you are using. Please revert to using pencil stubs and backs of used envelopes from the forgotten mail service. I am not a robot.

Eric said...

Your request is being considered. You raise some valid points in your comment. They are sharp points. One could injure oneself one time if one was not careful. The thing about good points is that they are always good. Good for future use and a very bright and sharp future. Thank you for your wise comment, Bike.