For us pros it actually a huge hinderance and does it help much at all often introducing more bugs than helping.
For things I am a noob at it is genuinely helpful to get me a leg up.

So it varies depending on your own knowledge base. It you need general guru level to not get tricked but an insidious hidden hallucination which will being down prod. I am Guru so not worried.

Noon developers will use ai and being down prod of not watched over.

However this is for non multiagent as most of the comments are. Multiagentic deployment is still in far infancy

I think the major issue is the use of the rule of 2 instead of using a better reading and weighting model when selecting which data points to return. That decision tree needs to be handled in a different way than standard ai and needs a different kind of algo to get better results. That is try e key item that needs to be worked on. The rule of 2 is causing the hallucinations.

They work for Cognizant. Cognizant is their employer. They can all themselves an Alphabet or google union but the only one that can fire them is Cognizant, not Google. Google can terminate the contract with Cognizant or ask them to leave, however but legally they need to negotiate with Cognizant.
By forming a union making it seem like they work for Google they are making a big complex mess for both employers, hence the firing makes total sense. If I was in charge and people formed a totally incomprehensible union instead of doing their jobs I would lay them off as well whether a leader in Google Or Cognizant. Workers are hired to do work, not go to city councils and complain. If they aren't working they can be easily fired.

When forming a union you need buyin from your employer and to work with them to establish the ground rules. And then things work very well, but if you form something weird and confusing expect to be axed.

We fired our CEO over one 0 that led to inflated reports. That kind of error also led to the firing of 90% of the rest of the staff because there was no cash to pay them. He should be promptly ejected. The staff like me who were actually profitable stayed on and kept things going for years after with a new competant CEO.

The article states:
      There are issues with eye drops that are causing harm to people over the last year.
      No Homeopathic products led to any harm to people over the last year.
      Because Homeopathic products do not have a legal requirment to be checked that can be faulty.

      So instead of regulating the root cause of the issue... they attack homeopathic products.
      A= true
      B= false
    if B do C
    aka they do C even though B is false.

      Exactly the same logic error with the Gaza conflict cause:
    Middle east leaders outsdie Israel treat all their citizens poorly.
    Israel is not treating any of the citizens outside it's country poorly.
    Middle east leaders outside Israeil say it's Israel's fault so we should support attacks against Israel.
    So they do C, support attacks against Israel while their own citizens suffer.
    like the FDA does C... le's regulate Homeopathy and not solve the root cause of the eye drop issues.

    Silly deflection at work again.... Homeopathy is not great, but why attack it and not the cause of the problem? Isreal is not great so attack it and not the cause of the problem....

EVs are charged from electricity, electirc power is mostly generated from oil and coal plants. Going from cars to EVs will just shift pumping oil to cars to pumping oil-> electiricty - cars and since there is loss in power transfer, countries with EVs will need import more oil to power their EVs so OPEC sees a net gain from EVs.

As we change oil/gas plants and increase wind/solar/nuclear/fusion and fracking .. OPEC hurts. Yes fracking... I don't like it but it hurts OPEC.

The figures look too human and elongated heads were a common practice in the area. So very unlikely these are aliens based on the info presented. DNA analysis should have been done 1st before showing these to people. fingers can fall off after all this time.
      We may get real live aliens soon and firmer evidence based on the recent hearings, but the odds of these being an alien is vanishingly low.
      These 2 are likely people and should have been left in their final resting places in Peru and not paraded about the world!

Worked on several Oracle deploys that succeeded and were on budget that I was on. It's neither Oracle nor SAP at fault here.

Any of these systems can be run much cheaper, safely and stably by hiring 1 or 2 good engineers. You try to go to these large vendors and larger projects and they will nip you in the behind everytime.

Must be run by MBAs trying to solve everything with powerpoints and sub standard architects and developers. Seen this many many times. It makes me cry.

Though eating less mean is better for the enviornment, it only affects short term global temperatures.
The production and consumption of meat has a net 0 impact on "long term" global warming (caused by new carbon increases to the oceans and atmosphere and water cycle). The cause of global warming is the burning and use of fossil fuels. Eating meat is the burining and usage of cows.
Just be aware average global temperatures has a net increase in the short term due to methane emissions from large animals, however and temperatures will go slighltly down after changing the global human diet, however the hockey stick graphs cause and the cause of global warming itself is fossil fuels.
By chance we use fossil fuels in the feeding of cattle and in meat production however we use the same fuels for everything else and only by eliminating their usage AND also by carbon capture can we return global temperatures to a sustainable range.

So, yes eating less meat is good for many reasons (it would take hours to list them all), however if we changed our diet the change to global temperatures would be minor, but not enough to undo the effects of global warming (long term). However in the shortterm we would see a very modest drop in global temperatures due to less methane in the atmosphere.

This is the famous, making an error in the effect of the taylor expansion in math error... it happens when you consider a small effect as significant.

That is correct, for most of them, not all.. there are a handful of exceptions... however....

Great, now swarms of unqualified people will stream into the US and mess up all of our IT systems even more until nothing functions even more so than today. We will pay them tons of rupees to destroy more technology is the name of Ram and "revert" us back to the partitions between storage systems until nothing works.

From the very start, hacking I have found pretty much trivial. It is our social responsability to not take down these computer systems even though it is easy to do so in order to keep the world secure. Writing a virus or exploiting a 0 day is very easy. This whole cyber security thing does 2 things. It prevents really dumb hackers from being able to hack. And it break existing real world code used to solve real world solutions. When a hack or virus is patched, what most don't realize is that act breaks all sorts of existing code used to provide end users with valuable code. At the same time things like PicoCTF are a ton of fun as you get to hack and are allowed to. White hat hacking is also a pleasure when I get a chance to.

I have seen the opposite, however I am just managing for 15 years with 5 years before that.

The best coders are the ones with advanced degrees who hold the company together. The ones who mess things are are the auto mechanic level hires who just break everything and keep saying.. yes I can fix it... while the systems fail for years and the managers look the other way except me.

I have never met a competant coder who did not have an advanced degree or at least a BA or BS. I know one PhD who did in fact argue for hours and got nothing done at all... just 1. You probably met him as well and then generalized to the whole population. His first name was Ryan.. a genius like me, but no he could not get things to work ever.

Employers either higher too many people who cannot do the job with or without college degrees. Of they interview and select capable candidates and keep those who do the work well with or without degrees.
    A degree is fine to require, depending on the position and to of course hire someone withour a degree if they fit the bill.
    Requiring a degree is good to lower the pool of candidates if you so choose.

      The real issue is IT. It is an art form that requires high capability to be done right, however recently it is being filled with hacks and fakes, for example the new CEO of Alphabet is running the company into the ground due to inexperience while Microsoft's CEO is doing his job like a rock star. Same with IT hires, sometimes rock stars are hired and other times a hack is hired. Solution is to fire the hacks before they drag your company to the great Pacific Garbage patch. I have seen so much stupidity lately in the name of IT, it is just a matter of time before the IT job cuts go from 10% to 80-90% to rid the industry of trash so it can thrive again.

    Degrees... bah.. that is not relevant... let's get to the meat of the matter and fix the industry before it implodes.

AI is using Wikipedia as one of it's main sources of training data most likely. The issue is if AI is trained on Wikipedia, and other things and the other things are less accurate and you feed that into wikipedia... AI will progressively get lower and lower quality training data and get worse and worse.

You would essentually start training AI on AI data and how would you weight what is higher and lower quality? The algos would spiral down and contradict themselves until things just don't work.

This is the classic chicken or egg contradiction playing out in a destructive fashion.

What people don't get is AI is just a glorified search engine. Use it to write the source and that will be happening and the little quality control we have with sites that wikipedia that strive to be accurate will just get worse and worse and quality will drop.

Wikipedia is essentially a source input.
AI is a source output.

If AI writes wikipedia you link input to output and you will generally get trouble.

Instead you can train AI by creating a new Wikipedia from AI and check how much it matches the old one and look for decreases in quality to both improve Wikipedia and impove the algos and get a win-win that way. But using it to generate Wikipedia and use it as an input, will be a way to spiral down to basically a scooby doo version of wikipedia with mysteries and hijinks galore.