My perspective is to figure out how to shape the way we use AI to show the right way to use it. This comes from a place of caring about the software and its users.
I strongly believe that when used properly, AI can remove a lot of overhead that comes with writing the code (from both a critical thinking standpoint and physically it actually takes a lot more time than people realize to manually type code). You can focus on the design and have your team much more integrated in the designs and specifications of things without it being a slog to document the damn thing.
If I can go off and write up a draft and get feedback from my team in a day instead of in a week, I think that's fine. We can collectively walk through and use everyone's domain knowledge to correct details. Once everyone is onboard with the spec, I can feed that into a model to progressively do small changes rapidly but in such a way that's still manageable and reviewable for myself and the rest of my team.
If I go off and merge 45000 lines of code into main: fucking fire me and my whole team that approved that.
As with all new things that have extreme views on either side (AI is complete shit versus AI is the second coming of Christ in this case), there's generally a middle ground in between that we can collectively correct toward.
Oh I agree! However, while they do brag about 30,000 PRs per day, they don't state what products these apply to.
I suspect for their public facing products (i.e., the claude ecosystem, which is like four products), the review process is much more rigorous, and the PRs merged in per day is likely much much lower.
A few things about their post: they're doing trunk development on a mono-repo. With their tool, Antique, instead of having branches, they just merge in stacks of commits, which they call "Change Requests". I have to wonder if their engineers have the models running and don't even realize the models are putting up a change request and if that's contributing to the high number.
Additionally, I would guess that a lot of the code merged into that mono repo is for experimental projects or other internal tooling that is in prototyping stage.
Marco Rubio has always had disdain for Trump, but appeased him to stay in office. He absolutely knew about this because word is coming out that on multiple occasions he and other representatives brought the suspicion about the GPU smuggling to Trump and Lutnick, but they did nothing.
Now that the midterms are over and Republicans don't have a majority, everyone is scrambling to spill the beans on everything to distance themselves from Trump in an attempt to save their careers in politics.
From the demo I saw, I actually really like this idea. I've thought about what generating UI on the fly for users with LLMs could look like, but my concern would always be wondering if the LLM was going to create a UI that's good, let alone usable?
So a few things that's got me a little fuzzy here with this protocol: how do I know the user is getting a UI that works? How do I test this? What kind of undertaking is it for a service to implement this protocol in their system (i.e., OpenAI, Anthropic, etc)? How do I know the model won't accidentally assign some critical functionality to the wrong button press in a generated UI?
And finally... my intuition is telling me, "oh like skills or mcp servers for embedding UI in the chat" - how far off is that thinking?
I work in manufacturing and would like to use LLMs for scenarios where there's obviously still rules for what should happen next, but there's just so many variables that I can't just code it into a decision tree easily.
As an example: I could have something I'm building fail. If I take a picture of it, can I see that it's
1) a false failure
2) salvageable
2a) decision-tree for *who* to send it to for reworking things
3) scrap
As I think of the above decision tree, I see some problems where I'm just asking myself, "why wouldn't I just use <insert a lot of options you can probably guess here>?".
I'll be honest, I think right now decision models are in a "solution looking for a problem stage", but I think there's opportunity still.
A reasoning cited in the article is that there was no source control in place for the trade tooling that Goldman Sachs finance employees were building.
I'm guessing that + probably the absence of any deployment infrastructure for the finance people to be forced to use made Goldman Sachs' eyes widen and pull the plug immediately before something blows up (if it hasn't already internally).
I also wonder if a lack of diligence with Claude Code (or any LLM tool for that matter) resulted in private information getting leaked, and that feedback is just now getting back to them.
In terms of the big picture of the industry and the AI bubble though, yeah this isn't too big of a deal.
If you're among the top 10% in the US, you're likely doing fine at the moment, but everyone else is struggling. On paper things may seem okay for the government and large companies reaping the benefits of this AI boom, but the current trajectory is not sustainable. Things will get ugly for everyone if we do not course correct on a number of issues like tariffs, Iran war, and more.
Gas prices are at an all time high, housing has never been less affordable, Intel stock just plunged a record 29% in its history, the US is now $40 trillion in debt... this is all just off the top of my head right now.
Inflation adjusted gas prices are pretty much average over last 25 years or so (and lower than they ever been in 2005-2014 outside of about 9 months' recession).
Inflation adjusted house prices are lower than 25 year average, and about as low as during early covid bottom. And housing affordability is almost best in history, with a few short stretches in late 1990s being better.
US debt load is by far lowest in history (or at least ever since reliable tracking appeared around 1963), for every dollar in debt Americans hold 8 dollars in assets. If you force people to pay all of their debt, at once (which is impossible), it won't cause a domino effect. Government debt? US is not a socialist country. Government can go to hell. And, all branches of government together are only slightly in red vs assets they hold. And debt to GDP is stabe in the last 5 years, last was the covid jump, which by now is about half undone.
Also, and here is a major thing that never existed before, consumption is now to a large degree disconnected from employment. Almost half of consumption is funded by capital, not labor. So potential for vicious cycle where people lose jobs, consume less, companies lose revenue, lay off people, and more people lose jobs, rinse and repeat, is now weaker than ever before - it is not impossible, but less likely. I think we underappreciate how valuable for economy this is. Domination of employment-derived consumption in the past has been a persistent source of instability.
So social negativity we observe is all just vibes, magnified by social media to stir the crowd. There is no money to make by making people feel happy - they click much harder if they are angry or frightened.
Sure there is a problem about a segment of population feeling displaced/irrelevant, but this is mostly a public messaging, and to a smaller extent, law enforcement problem.
Since serious crime is also at all-times low, i doubt it can escalate into something serious.
That's not the case really. People feel fucked because social media and politicians say them so to keep them engaged.
In fact, they are getting slightly better off. At the median.
https://fred.stlouisfed.org/series/LES1252881600Q#
check those charts, and then compare them to yours and you’ll see exactly what I mean.
wages have been roughly stagnant, while major life expenses have skyrocketed. but hey, if you don’t need medical care, don’t rent, and don’t need transportation, maybe it’s fine?
I really don't know what are the "medical expenses" that 90% of people under 65 need. Sure some get badly unlucky, but well, here in Europe majority of people under 50 just never visit a doctor, there is no need. And under 65, extremely rarely like a few times in 50-65 range. After that 65, you have medicare.
It's really perplexing why American 30-year olds cry about healthcare costs. I seriously don't know anyone under 45 who knows what it is. I brought my ex to the clinic once when she broke her toe, of course women visit gynecologist for checkup and to get pills prescribed once in 6 months, and that's it.
Here in Cyprus, healthcare works on 2.65% flat tax on all incomes, capped at 180K EUR income per year. Everyone seems happy and a family doctor makes 150K EUR per year, on par with the US earnings in a country which is much cheaper to live in.
Secret is much much much lower usage. People just use healthcare perhaps 10% as little as Americans do. It's just not in the culture. Life expectancy is still 83.5 years vs 79.0 in the USA. And yes we also have a car-based society with most people living in individual houses just like in the USA, almost no usable public transport, most of Cyprus is literally US-style suburbia, and in addition, everyone smokes.
So here is the advice of what to do with the healthcare: crush it. Same with universities, for the same reason. Many, many more people undergo unneeded medical interventions because of feeling compelled to do so for some cultural reasons, and many people also get useless degrees which they are unable to put to any good use, crazily jacking up prices for both healthcare and education.
Ironically, out of 3 you listed, two are flat in the last 15 years vs general CPI inflation, and one is slightly up. In terms of purchasing power of median wage, two are more accessible in 15 years and one (rents) very slightly up (~+9%).
Telling you: perception that median people are in some way struggling, is made up, it is political/social media tale they are selling you just to keep you angry and agitated. It does not reflect reality at all.
They may be struggling only in the sense of feeling that they are falling behind, and that is true, but this is an immutable, objective reality that does not depend on policy decisions, it is a force of technological change.
That's the thing: nominal wages are up more than 5x since 1980, even 2nd quartile https://fred.stlouisfed.org/series/LES1252881500Q. The chart i posted was for real wages, not nominal.
Result is that rent: slightly up, medical care: slightly up, transportation: slightly down, food: slightly down.
And sure, expenses that have a large component of human labor in them, especially qualified labor, skyrocketed. But others where automation plays a large role, fell. Overall inflation rate best matches expenses structure of the median person.
But why aren't they improving their lot A LOT, in line with the general economy? Well i am a coder and probably so are you. Why do you think job market, chances to get hired, and earning potential sucks for coders now? Who is to blame? Trump, Elon Musk, "late stage capitalism"? Who can we possibly protest against and what to demand? Nothing. It's just objective technological change, force of progress.
This. Consumption is up, savings rates are down. Core inputs into the economy are too expensive like meat and gas.
It’s only rosy in the aggregate and then maybe for areas overlooked and for the wealthy.
It can’t stay like this forever. Something will have to give. The rest of the working class folks and lower will not put up with being lied to all day long, well maybe if they voted Trump it won’t matter. Nvm nothing to see here. No minds ever change.
Yes that is a fair assessment. What are the downsides of it? More political idiocy, sure, that is a real one, as long as an average American can vote. Any other ones?
It doesn't mean some freemasons are grabbing all cash, or oligarchs. Most of the new wealth goes to "normal" rich and upper middle class. All of the top 10% are raking in big.
Seriously, i'm 47 and i only knew a SINGLE person in my life who escaped poverty through job. A lady who became a CFO and executive vice-president of a EU system-wide importance bank. And i don't hang out with poor people in general.
Maybe in US it's different, but here in Europe, "jobs" have always been merely a more or less guaranteed safety net to escape total destitution and becoming homeless. If you wanted to make it in life on even the most basic level (buy an apartment, get married etc.), you ALWAYS had to do something better than a job, or own assets.
Vast majority of jobs are not about economy, they are about social policy and politics (because people vote, so society has to at least somewhat placate them to avoid awfully nasty election outcomes).
I think Microsoft laying off 60,000 people was going to happen regardless of this AI boom or even the economy being in the absolute shitter it's in right now.
When the economy was much stronger in 2018 their headcount was around 131,000 people. By 2020 that jumped to 163,000, and by 2022 it reached 221,000. I have personally heard of employees feeling like they had nothing to do.
With their current head count at around 218,000 employees (I'm excluding their Activision head count here), bringing things back down to 160,000 or so employees sounds a bit more realistic, especially given that Microsoft has not been expanding its range of products significantly. They're building out more in the spaces they're in, but they've been pretty conservative on new product launches compared to others.
I strongly believe that when used properly, AI can remove a lot of overhead that comes with writing the code (from both a critical thinking standpoint and physically it actually takes a lot more time than people realize to manually type code). You can focus on the design and have your team much more integrated in the designs and specifications of things without it being a slog to document the damn thing.
If I can go off and write up a draft and get feedback from my team in a day instead of in a week, I think that's fine. We can collectively walk through and use everyone's domain knowledge to correct details. Once everyone is onboard with the spec, I can feed that into a model to progressively do small changes rapidly but in such a way that's still manageable and reviewable for myself and the rest of my team.
If I go off and merge 45000 lines of code into main: fucking fire me and my whole team that approved that.
As with all new things that have extreme views on either side (AI is complete shit versus AI is the second coming of Christ in this case), there's generally a middle ground in between that we can collectively correct toward.
reply