Everything posted by Vegas64
-
ChatGPT AI Tool— We all work for robots now
I know there was a time when the Grok model would release and "reset the market" with LLMs and be top tier, but I can honestly say I've never used it. Just not my world, I guess. I'm a ChatGPT and Gemini guy mainly with a sprinkling of Perplexity.
-
My new favorite subreddit
I'm mad vexed, it's what the projects made me. A rebel to the grain, there's no way to barricade me.
-
ChatGPT AI Tool— We all work for robots now
Maybe because I'm a neutral observer and AI is impersonal to me, you seem to think I'm a sunshine pumper or AI bull (check the stocks thread, I'm a noted bear) and think AI is a bubble.. But, I'm not sunshine pumping anything. I agree that the business model is busted. I think ads are a necessity at this point because the business model of a consumer technology like ChatGPT is busted. That said, I think we are conflating consumer AI and business or enterprise AI when you are talking about getting a useful work or product out of them. ChatGPT is a standalone consumer product today that can (and does) steal marketshare from search. Ads only help and I'll stop there as I'm just repeating myself probably. In the same way you think that I'm just some AI fun guy sunshine true believer, I'd challenge you to, what's the word here, extrospection? Look outside of yourself and your experience as a praticioner of AI in your narrow subset of your organization, role and responsibility. Widen your aperture, as we say, because whether we see it or recognize it, AI and AI strategy will have a seat in the board room for a long time. The business demand, whether the returns in the early innings are positive or not (and you are right to lampoon them in the short term as not, of course) is only growing. The demand for enterprise AI applications and use cases is not the same as blockchain or NFT or beanie babies or tulips. I bolster this opinion with the below, from TSMC's earnings yesterday which I thought super insightful. TSMC has been hedging on building out it's capacity to meet the overnight, hyperinflated, crazy boom and bubble of demand. The earnings have shown they have been conservative and cautious of spending on capex to build out foundries because in an AI bubble pop, they'd be left holding the bag if the demand evaporated. To quote him verbatim: The CEO of TSMC came out and essentially said as a business, we have to measure two or three times before cutting, because the risk is very high for us. I know you have an opinion based on where you sit, but can you acknowledge that the titans in this space might also have a perspective, maybe even some vision and understanding, that might lead to teh curation of a different opinion than yours (or perhaps even a nuance of the same opinion)? One that might inform a different strategy than you would suggest? One that might invite a discussion with some patience for those who might disagree slightly on some things, even while agreeing with a lot?
-
ChatGPT AI Tool— We all work for robots now
Revisiting the ad strategy for OpenAI. I read some interesting back and forth all day today I think I go back to ads being the obvious and needed strategy for OpenAI if they want to compete and win in a winner take most (and in which they have had a massive lead that they may have or may be squandering). A few points I've read and believe why: 1) In our economy any consumer internet product utilizes an adverstiing model because it optimizes for revenue income. You can increase average revenue per user without any worries about pricing/price elasticity because the entire increase in average rvenue isnt paid for by the consumer but rather by the advertiser who are willingly paying for it because they are getting a positive returns on their ad investments. This allows most to use it free and spreading the reach around the world (the Google and Social Media models). 2) Then when you have this virtuous cycle running, you are able to really full throttle on R&D, increasing advantages, because anyone without that model has orders of magnitude less because they have to either raise VC money, raise debt or charge users. 3) OpenAI's CFO basically admitted yesterday that OpenAI with it's fixed costs isn't a scalable business as it stands right now in that in order to make more money it needs to spend more money. 4) OpenAI needs to spend on R&D to maintain position and fend off competition. OpenAI's rapidly increasing costs match revenue almost 1:1. That's not a feasible company without an ad model long term, it doesn't seem like. Sam Altman was very opposed to ad models for OpenAI. The fact that cooler heads have prevailed say something.
-
I for one welcome our new AI overlords
Agree with all that. I would just say, regarding businseses being created and creating products and services that will be used and loved (or hated) by future us, based on technology that is emerging.
-
I for one welcome our new AI overlords
"run out of latent needs" seems like a static way of thinking about the complexities inherent in a dynamic world of ever-constant change. IMO.
-
I for one welcome our new AI overlords
For what it's worth, and I know there have been some pretty big misses (being too early is functionally the same as being wrong), but Big Tech has a pretty great hit rate for ignoring stated needs (and what people would not like) and instead building and delivering to latent needs that society didn't even know they would or could have. IMO that's the appeal of Silicon Valley for those who are idealistic and ambitious, before they are dulled and beaten down by money.
-
Greenland - Tassa Eller Asu !!!!
Thought this was an interesting view of it (politically):
-
ChatGPT AI Tool— We all work for robots now
Also, an easier habit to maybe start, is to simply add "-AI or -AR" to your search query in google. Or so I have heard. Add -ai to Query: Type your search followed by -ai (e.g., best laptops -ai) to exclude AI snippets.
-
Waymo cars Hit the Freeways
Thanks for the knowledge. I'd always conceptualized in my lizard brain "predatory pricing" to mean usurious pricing and taking advantage of people which inherently meant too high. I didn't consider it could be too low below a fair market value, in an effort to drive out competitors. I thought that was just good old fashioned business (leverage).
-
Greenland - Tassa Eller Asu !!!!
Not sure if we have a dedicated WEF Davos thread but with this being one of the biggest Davos in recent memory (shout out to Larry Fink), all the world's power players will be in Switzerland this week with Trump included. Deals get done in Davos, so maybe @Bateshorn your "ask for an insane amoutn of money" tactic gets used?
-
Trump’s America
I think this is spot on. It's why even so few "juniors" can succeed their pops in their nepo-baby way and make it happen and Don Jr. would never be able to fill the shoes. I think that's why you've seen the memes and murmers of Baron being groomed at a young age. Ultimately I think you nailed it, t's the Key Man risk with the ultimate cult of personality.
-
ChatGPT AI Tool— We all work for robots now
Thanks for this insight. Super interesting to hear the other side of it (practicioner, not marketing).
-
ChatGPT AI Tool— We all work for robots now
You don't think Agentic AI by way of creating digital workers is a good use case (or technically viable) in 2026? Asking because you obviously have a catbird seat and I'm just reading through earnings reports and technical pundit analysts and pattern matching/trend reports. That said, I agree and think everyone agrees actually, that we are in some sort of AI bubble. I would disagree (and would love to bookmark and come back and discuss in a year) that 2026 is the year the bubble pops. At least not in the way I think of bubble pops like dotcom, telecom, etc. before it and I don't think we see a stock market collapse due to an AI bubble pop. If anything, I think this is the year AI continues to diffuse and gets distributed more globally.
-
Indian
Yeah it feels like they really upped their prices once they got that Michelin star. As one does.
-
Indian
I know it's in the Galleria (bear with me here), but Musaafer was a treat. Super interesting food with a wonderfully built out ambiance and decor, a great variety of food from all parts of India and with a Michellin star for something or another with the original location in Tribeca. It's spendy for Indian food which is notoriously a great value (my party of 5 spent about $1200, lots of drinks and lots of food ordered family style) but the Onion Xuixo was the MVP. I had to go back to the well and order another.
-
ChatGPT AI Tool— We all work for robots now
Okay so let's take the position you are right because I don't know the details of the ad business model OpenAI is going to implement, but are you saying that ChatGPT adding an ad model is a bad idea? Or are you saying it's not going to matter in the long run as they aren't a viable and enduring business in the long run? I ask because Google is already a step ahead of the game in terms of monetization (AI overview, AI mode, etc.) and Google waking up last year from their slumber and getting into it gear was a Cod Red moment for Sam, well, this is an extension of that.
-
ChatGPT AI Tool— We all work for robots now
Adding Ads for ChatGPT reminds me of the adding of ads to Netflix. It changed the entire narrative and was a huge growth function for them and fundamentally changed the business and was something they should have done way before they actually did.
-
ChatGPT AI Tool— We all work for robots now
Bear with me here, but why WOULDN't adding ads not generate tens of billions in revenue? Saying that means that you think that ChatGPT will not attract the dollars and eyeballs of current social media today. Did you know that "Reels", a social media contraption widely seen as lame and culturally irrelevant (compared to TikTok, Youtube, etc.) brings in over $50 billion dollars in annual run rate business? If ChatGPT, advertisers potential dream from a user intent and call to action to buy standpoint, would do half of what Meta does that's tens of billions in revenue tomorrow.