Bites: A.I. and the inevitable death of the watch collector?

The biggest problem with using AI for suggestions is that watch collecting is an emotional pursuit. Logic plays a part, but emotions are what drives it.

If watch collecting was logical, nobody would buy an Omega for $10K when a $2K Longinesdoes the same thing, nobody would buy a $2K Longines, when a $1K Marathon is more accurate, etc...

I'd share my results as a laugh, but as mentioned, I strongly suspect many people would lie about their suggestions... As emotional apes we like to status signal, and having the "wrong" suggestions might signal low status.

I don't think there is a "perfect" watch for me(or anyone), I think it's more a case of the perfect watch at the moment.

Interesting discussion point, Kaysia. My answer is simple.

1) Preferences and tastes change. Priorities change. No one can predict the future. Life and human actions are equally unpredictable. That's the beauty of life and being human. We each have our own unique life situations and experiences. So, I am quite skeptical that AI can predict how I will choose my watch collection in the future.

2) AI is simply that, artificial. It's programming and algorithms are still created by humans. Even its supposed "self-learning" is programmed by software programmers. So, I will not contribute to the growth of AI by simply not using it. If no one uses it, it will become irrelevant. It's only hype to say that "AI will take over the world, and is the inevitable future." It will only lead to that if we bite the bait and succumb to the predictive mind programming. So, I will definitely not use any AI. Even if it uses my browsing history, etc., I will still choose to use my critical thinking faculties, and make my own decisions.

Congrats @JBird7986! Truly deserving. 🍻

Woah! I’m truly honored and completely humbled to have been selected as this week’s Photography Club winner! Thanks @BadgeHoarder and @Deeperblue!

Amazing shot. The reflection off the law books is captivating. 😜

@JBird7986

I appreciate the recognition Kaysia.

Keep up the Awsome posts!

Cheers! 🍻

I would not rely on the AI even if it was perfect in it's selection. The whole fun about watch collecting is the journey. Mistakes, patience, haggle, and success. Where is the fun in imputing data?

@Deeperblue Hello

The travel watch has been posted, what's the best way of letting people follow it and get attention for it, the actual watchcrunch guys are on board and are creating a badge for it. I've got a twitter account for it and my YouTube which I'm hoping people will send me videos of so I can post them. 👍

thank you Ben

Interesting thoughts!

So do you think your favourite watch changes with our mood?

Does this make the one watch collection impossible if we see that one watch as our perfect forever piece?

I guess the fact you've said that you feel there is no perfect watch already answers this!

Is a perfect three watch collection possible??

Are our collections destined to be in a constant state of flux???

(This would make me feel better about my current collecting habits 😆)

Ok, pretty much an unrelated tangent, but, thinking about this...

Preferences and tastes change. Priorities change. No one can predict the future. Life and human actions are equally unpredictable. That's the beauty of life and being human. We each have our own unique life situations and experiences. So, I am quite skeptical that AI can predict how I will choose my watch collection in the future.

In my opinion personality doesn't change unless there is physical or mental illness.

That is a constant throughout our lives and will steer much of our decisions. Yes, priorities will change, but what we love? .... Do you think what we love changes so easily?

You have won the world of watch collecting now. Congratulations 😁🎉🥳

So even if it told you your perfect watches, with 100% accuracy, you would still buy others anyway?

I do completely get that. A watch doesn't have to be perfect to be fun.

Ok, next question... Would you want to know your perfect 3? Or would you prefer to just discover them yourself?

Was an awesome choice 👌

That's great news!

Tag me in the posts when you put them up!

I am only one person and capturing all the posts for Bites each week relies on me trawling through the Latest tab every day, but stuff does get buried, so tag me if you need me to see something ☺

I am curious to see how it effects social media.

With deep fakes and CGI getting so good, I do wonder how long it will be before a YouTube channel comes out that is purely AI generated.

"Hey ChatGPT... create me 10 videos a day on watch content so I can make ad money with zero effort."

When AI can create the images, the videos and the content within minutes, it will leave us basic-b*tch humans in the dust.

At present, I think you can kinda tell when writing is AI generated... but it is getting better.

But as someone who still prefers an honest-to-goodness paper book as apposed to a kindle, it'll be a long time before I enjoy connecting with AI over an actual person.

I would like to discover the 3 myself only because you might come across something that was never in your wheel house and might perk your fancy.

Simplist answer I can give you is today's 3 might not be my 3 in the future. People's taste change not only due to age, but lifestyle, work, environment etc. Also, I think you would feel more satisfaction with the collection if you had to work to achieve it.

To hear different opinions and learn things I didn’t previously know.

I design Hal and he goes rogue, kills some astronauts in the process and proclaims the best 3 watch collection as a Hublot Big Bang, busted down Invicta and MVMT as a dress watch.

I asked Bing to create a 3-watch collection - they claim to have AI chat 4.0. Here is Bing’s AI response:

That’s a fun challenge! I found some articles online that offer some guidance on how to create the perfect 3-watch collection12. The general idea is to have a daily watch, a sports watch, and a dress watch. They should cover all your needs and preferences for different occasions and activities.

Of course, the specific models you choose depend on your personal taste, budget, and style. But here are some examples of possible combinations:

A Rolex Datejust for everyday wear, a Breitling Navitimer for sporty adventures, and an Omega De Ville for formal events.

A Seiko SKX007 for daily use, a Casio G-Shock for rugged activities, and a Nomos Tangente for elegant occasions.

A Tudor Black Bay 58 for versatile wear, a Garmin Fenix 6 for fitness tracking, and a Longines Master Collection for classic sophistication.

These are just some suggestions based on popular and well-regarded watches. You can mix and match any brands and models you like, as long as they fit your criteria and personality. The most important thing is to enjoy your collection and have fun with it! 😊

Received message. That's a fun challenge! I found some articles online that offer some guidance on how to create the perfect 3-watch collection . The general idea is to have a **daily watch**, a **sports watch**, and a **dress watch**. They should cover all your needs and preferences for different occasions and activities. Of course, the specific models you choose depend on your personal taste, budget, and style. But here are some examples of possible combinations: - A **Rolex Datejust** for everyday wear, a **Breitling Navitimer** for sporty adventures, and an **Omega De Ville** for formal events. - A **Seiko SKX007** for daily use, a **Casio G-Shock** for rugged activities, and a **Nomos Tangente** for elegant occasions. - A **Tudor Black Bay 58** for versatile wear, a **Garmin Fenix 6** for fitness tracking, and a **Longines Master Collection** for classic sophistication. These are just some suggestions based on popular and well-regarded watches. You can mix and match any brands and models you like, as long as they fit your criteria and personality. The most important thing is to enjoy your collection and have fun with it! 😊

In addition, Bing recommended Watch Crunch and Fratello as references for building a collection.

The conclusion from my completely unscientific study is that Bing May be a bad search engine, but at least it has good taste in watches…

What about thorium reactors? How's close are we on that? What do you think about the promise?

I've been testing AI for work. It's amazing for many applications.

The mistake I see many videos like these make us that they have no knowledge of prompt engineering. It's a rapidly evolving discipline, almost a programming language. There is multi prompting, tree of thought prompts, etc. The gist is that the more specifics you supply, the better the results. 99% of posts saying AI sucks is due to poor prompting.

And people who say it's a gimmick are underestimating AI. Yes, it generates text using statistical models but here's the thing. Words (and language) is how we construct models of reality.

Many people would be shocked at the results possible with precise prompting.

With all that said, I think the real problem in watch selection is the hands on experience. We all know that's night and day from pics. We need to solve that for many brands beyond the major houses.

The current AI engines are good at certain things and not so good at others. BTW I am working in Computer Science.

They are good at problem solving when the problem fits a pattern of previously encountered problems, and is objective rather than subjective.

The issue with subjectivity is not the emotional over the rational side, the problem is that the perfect solution of a subjective problem needs to take you into account, and the AI bots have very limited knowledge about you. They could give you generic advice about watches which then either is fairly objective but does not go very far, or advice that replicates common talking points of others. The result would be some kind of fashion barometer on watches. If a certain watch would be perfect for you, but hardly anyone talks about it then the chatbots won't talk about it either.

Outside chatbots, what we will get, and to some extent already got, is the gradual replacement of search&browse with the recommendations by an AI. This will be at the same time helpful and frustrating. Helpful, because it puts things you really want right in front of you, and frustrating, because meticilously researching what is out there to find that hidden gem you were never looking for in the first place will become more difficult. What complicates the matter is that these AIs can be subjected to outside influences, giving you less what you want to see and more what someone else want you to see: if you have been using the www for a while you may recall that google used to produce better search results than it does now.

I hate to give you a non-answer, but honestly I think it depends....

The problem with reactor technology development is that there are many outside forces influencing where money and effort go to. If it was only up to scientists we might spend all of our resources on researching the core of the technical challenge. Things like improving materials, optimizing the fuel and reactor physics, etc. If it were only up to organizations we would spend it on commercializing the tech to make an ROI as soon as possible. If it were only up to the public we'd spend it on analyzing every single safety and environmental case they can come up with. So doing all three at once, which is what ends up happening, means making it a reality ends up being a game of back and forth between all priorities.

Back to your question on thorium reactors. The science shows it works. The practical application, especially sourcing thorium or making it and materials with good enough properties to withstand things like molten salt, are the real barriers. I haven't even touched on codes and standards which are either very nacent or nonexistent.

I think the real practical approach is to go for the low hanging fruit first. To me that's recycling our current supply of used fuel and using in our existing reactors where possible. In parallel we put money into solving the scientific issues of new tech without the added pressure of commercialization. That way by the time we've gone through our used fuel supply the new tech is developed enough to actually be deployed practically.

There's also techs that are kinda in-between. Look up high temperature gas reactors (HTGR). They use a very special type of fuel called TRISO which itself is almost unmeltable. They could really work in industrial applications that use steam as a primary energy source. Again, with enough investment they may not be competitive with natural gas plants but they sure can work and sure will kill aloooot of the current GHG emissions out there if adopted.

Death of the watch collector?

Nope.

Addiction will prevail. 😜

So many questions in there not sure where to start so some random thoughts:

1. arbitrary limits like 'three watch collection' are just that - arbitrary and open to change. What if one day after getting the result of your perfect 3 watch collection you change the rules and go for four watches, or five or ten? Perhaps in going from 3 to 10 not a single one of your original 3 (likely to be fairly generic all rounders) make it onto the list of highly specialised watches for your 10 watch collection.

2. the nature of training AI will always make individual decisions tricky - the training requires huge population data that will tend to be a problem when reduced to an individual. Asking AI to find something "unique" still ends up in watches that are all over the internet - just less than rolex and tudor. This has a problem in time too - there will be more historic data than current so AI tends to lag a lot in places like fashion.

3. finally the AI will/may change us so that what the AI predicts is made obsolete by the AI. This is the marx problem. IE, marx predicts the workers will rise up and take over the state BUT capital owners see the prediction and change their behaviour to come to a different place and now marx looks like an idiot. you could see AI say - "you look like a perfect Invicta guy" and that then changes you.

Thank you!

In the mid-1990's I was heavily into an RPG called Shadowrun. "Where magic meets man and machine."

If you went the cyborg route, you were called a Street Samurai ("Street Sam" for short.)

Every character had an Essence score; an essence of 6 was fully human. An essence of 1 was cold, ruthless, and unfeeling. An essence of 0 meant death.

Most Street Sams had an essence of 0.62, which was pretty much the limit of how much cybernetics you could get without croaking.

With an essence below 1, you were particularly vulnerable to magic; because you were a whisker away from being an inanimate object. At that essence score, your humanity, your vibrance, your inspired heart, were pretty much things of the past.

In my opinion, the instant you turn yourself over to the influence of AI, you've begun the slippery slide down the slope of essence loss. The magic is gone. You are losing yourself. The machine is consuming you.

No thanks. I'll pass. 😎

Everyday, we are a different person, even in a miniscule way, either regressing or progressing, developing or stagnating, becoming better or becoming worse. Sure personalities may not seem to change in general, and it may steer our decisions. But imho, preferences and tastes always change to the degree our minds are open to change. Priorities change daily weekly monthly etc. depending on life situations, financial scarcity or abundance, physical well-being or sickness, etc. All these combine to influence how we decide to get a watch at present, and in the future. Add to this, we go through different seasons in life. There may be a season in one's life that one enjoys blingy stuff. Then after a decade or so, a person settles into a more simple aesthetic. Did that person's personality change? Or did preferences change? Or did circumstances change that influenced one's choices? Age can also influence ones choices. At a younger age, one might enjoy the sporty kinds of timepieces. Then, perhaps due to one's occupation or vocation, style preferences change. Today we may both appreciate a vintage piece, and perhaps we will pass it on to our children. But perhaps in 30 years, we may prefer a plastic cased quartz/ digital watch because it's lighter on wrist, and our eyes are having more difficulty seeing an analog piece.

All these to say that AI cannot predict one's future buying preferences so easily. Not if we feed it all the info it needs to decide for you. But, even the feeding of info can be trolled hehe...

P.S. My remarks are, of course, tongue-in-cheek, but the fact remains that I will never let an AI make my choices for me. 😎 It's bad enough that so much of my activity is chained to this stupid iPhone. 🤣 You kids get off my lawn!! 🤣

In my opinion, AI does not “create” anything. It can use a lot of existing data and analyze it with whatever mathematical formula you might like. If there isn’t data (obscure watches) you might get a data bias. Also, the algorithms can vary. I’ve found that they are ok for “weak” relationships but they may not have the availability to say something authoritatively (indicating a high confidence in the answer). I wouldn’t trust it to pick something for me.