100% agree with you. And I think the companies that figure out that you could take the same number of employees but have them using better tools are going to be the winners in the long wrong. Using AI to be more productive, shortening development cycles, etc is kind of the point, but real metrics and cost-benefit analysis needs to be done continually in any company adopting new technology. If it's not paying off, then don't do it. And I firmly believe that cutting large number of staff is the wrong optimization, being smaller is the opposite of growth. And in this industry, growth is success (in multiple dimensions)

Yea, there is a huge lack of understanding of what it's for. It's not a search engine or a program launcher. But on the other hand you can ask it (Glean, copilot, whatever) something like "Look for internal docs on XYZ and summarize. Note any discrepancies between sources. And cite your sources."

I find LLMs useful to make a workflow for doing some task. Like instead of asking it to generate tables, I have it generate a script that generates tables. Because I have on MANY occasions seen Claude Opus and others spit out tables of data with subtle errors in it. And while I usually have success when I ask it to review the tables it just generated, it has been much better at reviewing a small script's correctness than large tables of data.

Luckily I didn't claim it was replacing software engineers. Although it is a huge force multiplier for us. Docs, execution roadmaps, code reviews, security audits, benchmarking, trouble shooting, and writing boiler plate code. I think a good portion of my peers have been already using this tech for the last 2 years. We would have stopped using it if it didn't work at all, and it seems to be improving all the time.

For non software engineering jobs, which I think is where we're near the break even point on full employee replacement. Getting a system that can usefully answer help desk does exist, but it is expensive to run. Offering a inferior experience with a much cheaper AI is already possible and being done today.

Practically speaking. Getting one experience QA engineer to direct an army of agents at directed and ad hoc testing is already happening. So those entry level QA jobs that many of us used to get our foot in the door into the software industry are not going to be available (sending those jobs over seas already made them an unreliable route for most of us in the West)

The only thing slowing down adoption of AI is that usable AI costs slightly more than a full-time employee. In 2026 there is no cost savings in replacing your work force with ML/AI systems, although AI can be a work multiplier in some industries allowing for increases in productivity. We're already seeing the cooling off on the hiring of new college grads for engineering and tech. An old timer can steer several agent projects at once versus training up a junior software developer. The immediate benefits are obvious, but long-term it's a disaster. Because I can only babysit so many agents at once, and I'll be walking out of the industry soon, with conceivably fewer people to take my place.

As for the current layoffs? Nothing to do with AI at all. No increase in profit is profitable if you layoff your staff and "replace" them with AI, because the numbers simply don't add up. Maybe in a year or two it adds up. But right now, we're seeing companies cutting reoccurring expenses (like payroll) in order to prepare for economic downturn or recession. The tech industry is going into a stealth sleep mode, we've painted eyes onto our eyelids or put on dark sunglasses, and the rest of the class still hasn't caught on that we're taking a nap.

AI will be running finance too, so aim for manual skills that can feed your family or work in a barter economy.

Your right to operate a business based on lies is not constitutionally protected. Of course our judiciary has been sucking capitalist cock for so long that they are no longer capable of impartiality.

And MS should be forced to allow activation of XP. If I have a legal copy then I should be able to install it. Microsoft is the one that coded up the hoops we had to jump through, so they ought to be liable for supporting that process indefinitely or remove that activation behavior.

It's also toxic if you ingest or inhale or touch it. People are used to the risks of gasoline, or complacent. Nice thing about lithium batteries is that you won't have to inhale the fumes every morning while stuck in traffic. (Only solution to traffic jams is bikes, buses, and trains. EVs won't solve that quality of life issue)

My tablet is an N350 (Star Labs StarLite), roughly in the same ballpark as N100 and N150. It's enough to play on dndbeyond. But whatever you considered "very capable" does not align what I consider barely capable.

I stand by my OP that it's a piece of crap. And that it's not about AI but about getting compromised hardware into your home.

Except it's too low spec to play games or do any heavy browsing. So it becomes a foot-in-the-door for an AI agent to snoop your home networks and copy your personal information. For the low low price of $399. Plus whatever you will need to pay to Anthropic, OpenAI, etc to actually have access to their APIs when free tiers disappear next year.

— Rudyard Kipling (1865-1936), Dane-geld, Stanzas 5-6

There is a good reason that law "law enforcement agencies around the world" advise again paying cyber criminals. And it isn't because law enforcement is dumb, or that they like seeing you getting your data stolen.

I was one of the lucky ones to get a Google CR-48 prototype back in 2015, and at the time, it was simply amazing. Yes, it had its shortcomings, but damn, it was impressive. While today's Chromebooks/Googlebooks sometimes feel like a marketing team collectively threw up on them, they have come a long way. My hope is that there will be enough settings on the new devices to tweak it to my liking.

I remember perhaps 20 years ago, some futurists were predicting that most of the time we'd interact with agents. That we'd each have our own personal assistant to curate and present us the information we need. This particular interview (sorry, I don't remember who it was) used the example that an agent would put together your daily newspaper specifically for you from multiple sources. I thought it seemed like kind of a stupid idea...

What Digg and others don't realize is that websites will be dead in 10 years. At least sites that people regularly visit for content. Nobody will visit them directly. Everything will be curated behind app stores and AI will present summaries. Even YouTube isn't safe.

Between microphones and acoustic fiber optics, it's getting so that concerned citizens can no longer talk about their concerns.

I would like my new Trump Phone before the economy crashes. But luckily we have more than six weeks before we have to worry about any serious issues with our economy.