How much experience do you have in the field? If it's a lot, aren't you worried about your own job security?
Something that I actually agree with. I'm only gonna start worrying about my job if I start taking orders from a machine.
Artificial Intelligence thread
- Thread starter 9dashline
- Start date
They also promised an SLM thats omnimodal... thats gonna be a game charger in robotics and physical AI which is the next big thing and perhaps final stretch to AGI....
"Open"AI said they won't even show the users the chain of thought process of their o1 model...
If Alibaba dumps the weights to its q1, thats gonna be a punch in Sam Altmans teeth
Ask him about EROEI.
I have been hearing that Software Engineers will be out of a job because of this or that magic piece of software for like 30 years already.
So far this AI is just another better form of automatic code generation. It will reduce the typing effort for writing software, but little of the mental effort. You still need to talk to the customer, draw up specifications, and think of how to solve the problem by yourself. You ask the AI to generate the code you desire, but then you have to double and triple check it to see if it does what you want, and if it doesn't have bugs.
Like I said, people underestimate the effort of Software Engineering. The moment AI replaces Software Engineers, basically all white collar jobs will be obsolete, from people working in banking and insurance, to engineers designing engines, vehicles, you name it.
it's almost impossible trying to explain to some people how hard it would be too automate away high end software engineering at this point.
AI is going to drive a lot of changes in our work place. I'm convinced of it. The work I'm doing right now is probably going to put some people out of work. But adoption of technology takes time. It often takes disrupters to enter industries to really cause changes.
BYD is probably leading the effort for putting AI robots to work. Even it is only estimating 20 to 30% of workers being replaced by robots in the coming years.
We just haven't seen what it looks like for robots to do works that researchers do yet.
yes, I read this entry and posted it already. All this tells me is that Google has an advantage over competition.
The entire idea of using national network to do computation is not new.
That's why China's EDWC project is all about.
It’s because they haven’t worked with large software in production and just assumed that it is like small self contained school projects.
Seriously whenever I hear this crap I wonder what environment they're working in. I have been using GPT 4, Co-pilot etc. tailored to our (large codebase) systems and it just produces slop all the time. Useful slop, yes, but slop nonetheless.
It makes it so I don't have to break out regex as often and reduces boiler plate, but that's something compilers have been doing since forever. It's also fundamentally reliant on problems being common. But common coding was never difficult or time-consuming to begin with. It's always been obscure bugs, constraints that are hidden in 1 commit in a 10,000 of a public library, dealing with obtuse management requests, etc. that consume the most time and energy and that's not something LLMs help with.
Going from 3 to 3.5 to 4 has improved the quality of the slop but hasn't tackled any of the other issues that taking developer time. The limitations are baked in to the way LLMs work.
Give it two more years...
Coding's days are numbered... the tokens are embedded into the walls so to speak...
Last edited:
I can give it twenty years and AI wouldn't replace coding or programmers - at the end of the day you need someone to actually look through, test and debug the output, which requires someone knowledgable in writing and developing software.
If an AI is capable of self developing, debugging and testing software, it must have the ability to be self aware and actually understand what it's creating. Therefore it has certainly ascended into an AGI, at which point we have bigger problems.