steveBK123

Right the most likely outcome to me is a pop and slow fizzle. Mega capex labs become uneconomic at the prices they need to charge to do more training runs. Non frontier cheap models become good enough at far less cost that people use that.

The use cases LLMs are useful for without lots of RF/fine tuning proves more limited than the expansive views currently held.

We get a generalized productivity boost the way the internet gave us, without any one sector or company being a winner take all of that productivity gain.

Agreed but the AI lab investors and CEOs aren’t ideologically aligned with concepts like UBI. Some of them have even commented on social media about the unsustainability of social security and need for cuts. Also why would you they do trillions in capex investments just to give their returns away.

They are clearly smart/dumb enough to think pushing labor replacement to 20%+++ unemployment levels would have no consequences for them personally. Maybe with their private jets, private islands, doomsday shelters, etc they are right.

Seems like a bit of a gamble to me, or probably anyone who has read a bit of history. But I’m not a billionaire.

The challenge with cloud has always been the selling point of “infinite scaling” vs the flip side “infinite billing”.

It’s not in their interest to make cost controls work well.

Ideally you’d be able to set something granular like “allow this service to scale up only 10x, measured at an hourly level, and alert me when it happens. Drop all requests that exceed 10x”.

And then you are mostly in a throttling situation until the burst clears or a human can review & accept increased usage is ok/increase thresholds. I’d rather have services go slow during excess load (like a real server) than go dark for remainder of month.

This seems a lot better than brute force “turn everything off at $X level of monthly billing” or “no limits you can charge me infinity dollars”.

Extremely sarcastic take but - shouldn't the 10x productivity boost of LLMs like their own in-house Gemini make maintaining software easier?

Nothing in terms of feature release / quality / maintenance I am seeing out Big Software would lead me to believe there's been any productivity increase.

I don't even think this is an extrovert/intropvert thing completely, so much as what people are used to/how they were brought up.

Some people are very concerned with appearances and therefore uncomfortable with people in their homes (in either direction - too shabby, or flaunting).

Other people (with some overlap with above) are too OCD? about planning and so they turn hosting into such an ordeal they are overwhelmed by it.

In general I find that while I am an introvert, hosting a group isn't really a big deal. I find 2:2 with another couple to be more draining than a group of 10 in terms of being able to step away and catch a moment of calm. Further, as a host .. I get to select who is coming in the first place!

Agreed an enterprise guy is not the next CEO but I doubt Zuck has any near term succession plan.

Mosseri is basically the same age as Zuck, and Zuck still seems extremely hands on.

Check back in 10 years to see how things are going, maybe its Wang or a currently unknown 20 something.

Ivory tower?

The problem with all implementations so far is they are slow, verbose, and clunky.

People working on these projects should study pre-digital forms of non-written communication. Commodities exchange pit floor hand signals. Pilot-ATC comms. Military radio. Etc.

If I want Audio AR, I want it to be like a co-pilot flying a jet with me. Crisp, concise, zero fluff.

Make it a slider for people who want to get chatted up vs those that don't.

> How does someone write software that needs to run for ~50 years

You in a lot of ways have to respect the ability to have done this in the first place, compared to todays slopware which manages to break within 1-2 mandatory OS update cycles.

More things in memory also leads to making more connections / recognizing more types of patterns more easily.

There are things that you just see/know/recognize that are not easily searchable/promptable because the fact that you should be searching/prompting would not occur if you had not seen the pattern before.

One could argue "experience" is just the accumulation of memory such that you can recognize more patterns more rapidly, short circuiting bad decisions before they are made. With LLMs its even more important to have a well tuned "what should be getting built" mental framework if that becomes our primary job.

You know its interesting, the physical thermostat has more settings than the app.

I couldn't find a "disable Spotify" setting, only disable Alexa (already was) and to turn the speaker down to zero (did that).

How did you disable?

> Just about the only thing I might want to see on a smart fridge door is maybe a shared family calendar that can sync with an external calendar provider, and that's not even listed as a feature.

Fundamentally due to product lifecycle expectations, I wouldn't even want that!

I hope a fridge will last 10+ years.

Unfortunately any SaaS/cloud-connected piece of hardware is going to have some combo of firmware/OS/security/app update deprecation well before 10 years is up. The last people I expect to maintain software that long is the maker of a fridge.

There's so many companies reputation-tarnishing themselves by doing this "we sell a lesser line thats actually terrible" thing. Why do you think I will now trust you with more money for an upgraded model? You've just made me skeptical of your brand entirely.

I've sworn off Bosch appliances after a 2 year fiasco of service calls with one of their dishwashers. The thing was still about $900 plus install. At what price point am I to believe it's not the worse-on-purpose line, as they cheapen the product every year. They seem to have models from $700 to $2000.

A bad $900 model sours me on gambling that the $1200, $1500 or $2000 model is the good one this year.

Sure but the infotainment/smart BS should be firewall off from the "keep the food cold" life safety part of the appliance.

Even Tesla which plays fast & loose with some things.. the car does not come to a halt if the infotainment/GPS/etc tablet reboots. It's VERY disconcerting as you have no speedometer, the car is eerily silent, etc.. but the car still drives.

Just sloppy, cheap, lazy engineering.

Wildly its an Ecobee and they added the feature on purpose some years back.

Apparently a firewall rule is about the only way to stop this.

There is SO much of this in IoT / smart / 2010s-2020s tech.

Recent household BS: music playing out of my thermostat.

Wife tried playing music out of Sonos, but the clunky app was laggy and had also signed her out of Spotify. No problem, you can also just play direct from Spotify app into Sonos speakers. She opens Spotify, selects "Living Room" and next thing you know our THERMOSTAT was playing music.

Neither of us know this was possible, neither of us enabled it, apparently it's existed for years but for some reason Saturday was the day that "Living Room" thermostat beat out "Living Room" Sonos in the Spotify sort order.

There were product managers on both sides of that decision to prioritize enabling such a stupid feature.

I'd pay for it to not be possible. I'd disable it on either/both ends if they had a toggle, which of course they don't. MVP baby.

Besides Samsung incompetence, I think smart fridges were probably solving the right problem (some sense of efficiency / time saving in grocery management) the wrong way (sensors/cameras/connected fridge).

It turns out that grocery stores with real time inventory on websites & pickup/delivery are a much better solution.

I have my weekly order I add my usuals to, and then add items to the cart all week as I noticed I am out/low while making dinner.

This is the perfect story of 2020s era tech.

Just total engineering incompetence of forced updates with new features for no reason that degrade core workflows and introduce needless instability.

Claiming they got all employee & spouse data.

Wondering about pets..

Depends on if you are buying tokens or being paid for tokens

TAM: $50T

We've invented the opposite of lossless compression

Defending the same entities getting large DoD contracts to use AI for killing?

Well theres at least two different buckets of this.

First is the scraping of the open internet.

The second is the paywall bypassing, YouTube audio recording, and pirated content training that the labs have basically admitted to in one form or another.

Content from both gets served back to us, in exchange for watching ads/paying a subscription/paying tokens.

The second is more immediately hypocritical because they are license/copyright/DMCA violations that the little guy could get sued for while the labs get $2T valuations for. The automation of crime at scale, which is a common VC pattern.

I suspect at this point, given the nature of the AWS org.. they actually don't even know themselves either.

Slop blog

First of all they could write an amazing slop blog PR post about it afterwards

You'd be surprised how many customers are lead to believe this by a combination of opaque marketing, aggressive sales, and willful ignorance.

I worked at a 2000 person shop where the CTO moved us from 1 on-prem dc (we were begging to add a secondary site for years due to outage risk) to 1 AWS region.

It's not exactly straightforward to run multi-region across an alphabet soup of AWS services without decent configuration / application awareness, paying at least double.

There's something reassuring in this for me.

There's a lot of magic & handwaving from hyperscalers like AWS about redundancy. I always wondered about some of the engineering to make this absolutely (and literally) bullet proof. At the end of the day most of their answers when you push hard enough involved paying 2-3x to run everything across multiple zones/regions, and lots of awareness in your application to handle this.

In any case, I think it's good that when a data center blows up the data is lost. Noteworthy for future skynet situation, etc.

A lot of this stuff again seems like labor augmentation not replacement, like in this article. She wasn't using professional paid artists previously, but a mix of her own extra labor (which she had less time for with a baby) & a barista who drew well (but had recently left). Most small businesses are like this.

I'm not an AI booster, but I think GenZ particularly have really lost their mind with the anti-AI rage.

I was handing out flyers at a fundraiser, for a charity to promote their next event. The flyers were made by a staff member using some AI art.

Some GenZ kid started ranting to me about how the flyers were AI slop and how some artist would have been willing to do the job, blah blah blah.

Kid, its a charity event that I'm volunteering for on a beautiful sunny summer weekend day for.. it's gonna raise like $10k if we are lucky.. no artist was going to paid with or without AI.