In still on AM4 and honestly see no reason to upgrade. I want more ram is all

Anything that filters through the emotional part of someone brain before the logical one would work.

If you are using an agent that doesnt have an approval step before applying changes, you deserve this. You werent even reading the code being produced.

Water is a universal human right, except when it involves people I dislike. Thats what you sound like

Its here to stay, but we are in the “we dont know the best way to use it try everything” phase.

From my experience using it as a consumer, and writing custom agents; i think the way it will pan out is as specific tooling that unlocks things that weren reaaonable to do before. Key word here is tool. Its just a god damned tool.

We will still need to hire people, because something llms can never do is be held accountable. Someone needs to use the tool.

As far as the data centers, this is an overenthusastic bet that all compute will be rented. It wont. At best a few additional datacenters needed for hosted inference, not at the scale they are going for.

A board with a nail in it!

I thunk AI can be great. Has helped me for sure. Its just how big tech wants total control of it, and the compute necessary to run it.

Best case scenario atm is competitve Chinese ram and gpus come on the market and are “good enough”. Atm the chinese open source models are nearly as good, and quite honestly good enough for many tasks.

We start by taxing shares used as collateral for loans. We need to disincentivize that trick that allows them to pay lower taxes by not actually selling shares. If it can be used as collateral, it has a value and should be taxed as if it were a sale of the shares.

When that happens, its imperitive that you dont participate. We need to reuse, and repair current and older machines.

Yes, but votes are proportional to the number of shares owned, and most of the shares owned by non insiders are owned via index funds. So you meed to account for shares owned by the board members (including Elon himself), in addition to shares owned via proxy (shares in index funds, etfs, etc) where the proxy just votes with managment’s guidance. Its not like voting for a politician. Its more akin to 77% of shares voted for it, which translates to a small number of people. The remining 23% are likely individual shareholders, hedge funds and family offices that disagree with the compensation.

Its more that huge firms like vanguard and fidellity just default to what managment votes for; they dont meddle. The individuals holding the shares of the index funds likey did not support it.

If you want total control you need to take 100% of the risk. If you want to use other peoples money then you are gonna have a board. Typical elon wants to have his cake and eat it too.

If you asked me last year, i might say its an existential crisis and they wont come back. Now though, I dont think thats necessarily a guarenteed outcome.

The more I use AI at work, and the more I experiment with custom harnesses/agents, the more clear it is that every task it does provides an output that must be validated and checked. Therefore, its a convenience tool.

Before VisiCalc, peoole had to make spreadsheets by hand on litteral paper. Manual calculations with a seperate calculator. Did VisiCalc (and later Excel) get rid of entry level jobs? I dont think it did.

We already have ample evidence of large companies trying to replace people with AI, with disasterous results. For the companies that have been reducing headcount because of AI, all they are doing is exposing themselves to future competitors. MS and Google become IBM and Bell Labs. New, hungry competitora take their place, and they hire like mad so their competitors dont get the talent.

We are seeing this play out right now in the case of hardware. Buying up all current and future supply to prevent competitors from having enough compute (without them getting a taste of course), but thats another story.

As to when? Beats me man, i work for a small startup that is still in “hire people we know” mode.

Im still not totally convinced that its going to take entry level CS jobs. I have the concern, but part of me thinks the job will expand to things like MCP server building, harness creation or implementation, etc. In other words, any jobs lost to llm efficiency in producing code will be offset by needing to build more stuff.

In addition, even if agents are doing some of the gruntwork, they wont be fully autonomous as they currently are. Anyone doing that (unsupervised coding) is going to get burned. Therefore, the new bottlenecks is review, but also business contezt complexity.

Can i build things faater than before? Yes, but my new problem is keeping track of it all.

I would kill for a ressurected battlefield 2142. I know there is a community one out there but I wish it made a comeback as a remaster or something.

Thats ezactly what they want to do, bur when that happens we must resist it. Play old game. Use legacy hardware. Participation is tantamount to acceptance.

They were usually just friend groups who took their entire computers over to someones house.

Think sleepover but you also bring a box of electronics

Its just too fishy that there is always, conveniently, a singlr dem vote in the senate to support right wing bullshit.

Worker coops! The only way to get that done is to statt a company with your own money so that you dont need to answer to a board/investors

I think thats completly false. Llm and be held accountable like a manager can.

The real danger imo is that hiring entry level needs to be deliberate. We MUST train the next generation and provide oppurtunity.

We will hire then, they will use AI, and it will bite them in the ass. This is a good thing though becauae we learn by getting burned.

Im ignoring the data center issue, which is really a “we wanna make money from subscriptions” scam. But open source models running on local hardware will sort that out over time.