361–370 of 394 posts
Earlier quoted context omitted.
The R&D tax credit change actually took effect in 2022, and one of the few good things Trump's BBB did was reverse it
It only reversed it for within the US, I learned that when the company I worked for (owner was a US company) closed.
If you want to do R&D overseas, best to set up an overseas company.
Why would remote working have anything to do with the amount of junior hiring? You can maybe make the case that remote work isn't as helpful for juniors etc etc. but that has absolutely nothing to do with the decision to create a job requisition or not.
Earlier quoted context omitted.
I don't think that part is true, either. The average human could be trained to use an agent to synthesize information in their job to help make product decisions. The average human could not be trained to evaluate whether a reasoning model produced a correct proof in research-level mathematics. To be sure: reviewing a candidate proof at this level written by AI is significantly easier and faster than writing and crea…
I didn’t say any of that either, all I said was for ai, one type of task is easier than another I said nothing about a human’s capabilities, which are most certainly different than an AI’s This is the second time you’ve said a bunch of stuff I didn’t say, and then tried to argue with me over it.
Earlier quoted context omitted.
But most of the stuff it returns existed before No, it did not. Start with clean llm, no external previous ideas of humans inserted into it, and let it generate some wisdom on its own and then lets talk. An LLM is about as likely to do that as you are. The ability to generate "wisdom" ab initio cannot possibly be a criterion for intelligent reasoning. The ability to arrive at novel mathematics proofs, on the other ha…
>No, it did not. It did. In the form of a pattern that humans are too incapable of recongizing. LLMs identify and repeat the pattern. That is all. For example, if A -> B and B -> C, logic dictates A -> C. But LLMs will be able to state that A -> C, without actually using logic, if there is sufficient statements in its training data that says A -> B and B -> C and A -> C. So now if you say P -> Q and Q -> R, it will s…
You're not feeling even the faintest stirrings of cognitive dissonance when a "text search" solves math problems you couldn't solve yourself?
Earlier quoted context omitted.
But not logs or code. Please do not send me screenshots of logs or code.
A phone's font is large. A picture let's you see code/logs properly formatted and in proper context. It also gives you the opportunity to circle/arrow and otherwise annotate the points of interest while displaying context around. A picture prevents you from copying, but often that's not as important. One can use judgment to determine whether a picture or text can be better. Receiving such messages can sometimes mean…
I tend to get things like 1/3 of a terminal window missing all of the context. No commands and only half of the error.
Ever receive a photo (not a screenshot) of a spreadsheet when they expect you to do something with the content? I have.
Another personal favourite is a browser window cropped so tight I can't tell where they are. No address bar, page header removed, sidebar missing, etc.
Send text as text! One of these times I am going to print the photo, annotate it, and fax it back. Or maybe I need better coworkers.
Earlier quoted context omitted.
A phone's font is large. A picture let's you see code/logs properly formatted and in proper context. It also gives you the opportunity to circle/arrow and otherwise annotate the points of interest while displaying context around. A picture prevents you from copying, but often that's not as important. One can use judgment to determine whether a picture or text can be better. Receiving such messages can sometimes mean…
I don't think people are using any judgement, but if they are it is only poor judgement. I've rarely received a helpful screenshot like you describe. I tend to get things like 1/3 of a terminal window missing all of the context. No commands and only half of the error. Ever receive a photo (not a screenshot) of a spreadsheet when they expect you to do something with the content? I have. Another personal favourite is a…
Earlier quoted context omitted.
> Social engineering won't work on a human security expert who knows and understands the implications of the information they are giving away Social engineering, like prompt injection, is a context attack — easy to spot if you're ready for it, but harder in different circumstances (rushed, panicked, tired, having a bad day, etc.). Troy Hunt (security consultant, creator of HaveIBeenPwned) and Cory Doctorow have both…
> easy to spot if you're ready for it Are you serious? LLMs, being a computer program, should always "be ready"? Unless you want to also claim that LLMs can be rushed, panicked, tired or can have a "bad day"! Jesus! The mental gymnastics people will go through to justify LLMs is just absurd!
Largely agree with you otherwise, not sure why you read my comment as mental gymnastics to justify LLMs. I don't think that they have an internal emotional state that can feel rushed, panicked, so on. They do — superficially — "respond" to such cues in language, which is why they can be "threatened" [0] and "flattered" [1]. But without an internal theory of mind, LLMs do this sycophantically without any internal model of the world (hence the quotes above, to avoid fully anthropomorphizing their behavior).
The only parallel I'm drawing is that both humans and LLMs can be coerced into unintended behavior via language ("social engineering" and "prompt jailbreaking" respectively), and those attacks are more effective if an attacker is allowed to control more "context", even if the underlying mechanism of why those attacks work is completely different.
[0] https://arxiv.org/pdf/2507.21133
[1] https://arstechnica.com/science/2025/09/these-psychological-...
Earlier quoted context omitted.
My first and only layoff was effective 1h before that law went into effect. 1,000s of us were shunted off b/c an entire research arm was canned due to the changing cost of research s/w teams. I'm poorer but happier now b/c of it. That job was nuts.
I took a decent sized pay cut about 6 years ago to join a small company and build out our own product knowing that I may not get a raise for awhile but got to work on something cool with people I liked and I haven't regreted it for a moment. Looking forward to waking up to do work is a wild feeling.
Earlier quoted context omitted.
This is a fair criticism and I am not defending the practice. My understanding is that time-bombs like this are very regularly introduced into all sorts of bills, party-agnostic. It's how they can say things like "We saved $X over Y years!" where a lot of the time-bombs go off half-way through the 'Y-years" bit unless renewed. Please correct if I'm wrong about this. I only know what I read, which is hard to trust any…
They’ve been regularly introduced in a party-agnostic way, exclusively by republicans. But yea, “party agnostic”.
What if there are several reasons all at once? * Economy is tighter overall. * Covid overhiring followed by firing means fresh graduates face competition from a large cohort of people with a few years of work experience. * Even if it's not necessarily true, if the C-suite _believes_ that AI can replace juniors, that's enough to funnel money away from hiring and into AI investments. When money for hiring is tight, sen…
Places are still hiring juniors, but now the bar for them is just much much higher, because a large portion of the junior level work (i.e, execution on implementation details), is pretty much 80-90% done by some senior + AI at this point, whether people believe it or not. This was certainly the case at the FAANG I was at.
I don't know why this "COVID overhiring" rhetoric is still a talking point. Covid was over 5 years ago at this point, those "overhires" have already left or have been flushed out from all these layoffs. Are you telling me the "COVID" overhiring resulted in such a huge surplus that, 5-6 years later, they are still on the market?