* Posts by ericsmith881

12 publicly visible posts • joined 10 Aug 2020

Researchers made an OpenAI GPT-3 medical chatbot as an experiment. It told a mock patient to kill themselves

ericsmith881

You forgot the absolute requirement of being practically unable to speak the languages commonly used by people calling into the call center. Uninformative, unintelligent, unhelpful, AND incomprehensible are listed right on the job description.

ericsmith881

Re: Very artificial intelligence

Depends. GPT-3, at its current level of sophistication, is about on par with a toddler. If you have a job a toddler could do, you would pay and retain said toddler (or GPT-3 equivalent) to do it.

This is not an insult to GPT-3 or toddlers. It is simply the realization that such technologies are very experimental at this time. Nobody is born knowing quantum mechanics. You have to learn to walk, run, speak, read, addition, subtraction, multiplication, division, calculus, differential equations, classical physics, special relativity, general relativity, and THEN you begin to understand quantum mechanics. That takes decades for the average human. Give the AI a break. It's a toddler. It will grow and learn and get better, and we humans will get better at teaching it at the same time.

ericsmith881

Re: Curious

To be fair, nobody ever claimed GPT-3 was ready for this sort of thing. If you push anything beyond its limits, failure is the *expected* outcome and should not be a surprise to anyone. If you designed a bridge to hold 10 tons and it fails when you put a 20 ton load on it, do all bridges suddenly fall into disrepute?

This stuff is in its infancy. That it functions at all is a miracle. It will get better over time. in the meantime, use it for what it is: an experimental tool with limitations.

Yes, it's down again: Microsoft's Office 365 takes yet another mid-week tumble, Azure also unwell

ericsmith881

For all the companies rushing headlong into putting absolutely anything and everything into the cloud, this *should* serve as cause for alarm. Of course, it won't, but it should. Cloud does not equal infallible. Too many companies are putting all their eggs into one basket. This has always been a bad strategy and it remains so even in the cloud era.

Another reminder that bias, testing, diversity is needed in machine learning: Twitter's image-crop AI may favor white men, women's chests

ericsmith881

Seriously?

Oh for crying out loud...the issue here is one of getting proper contrast for facial recognition. Anyone who's done facial recognition knows this. The whole "pale globe" comment is ridiculous. The picture in question shows a dark-skinned man in a very backlit situation, something difficult for *any* algorithm to lock onto. You'd get the same results if a white person was silhouetted. If it can't detect facial features due to poor lighting, it's going to lock onto anything that remotely resembles a face. It's not "choosing" a pale globe because it hates black people.

Yet another example of people getting offended about something simply because they're looking for reasons to get offended.

Engineer admits he wiped 456 Cisco WebEx VMs from AWS after leaving the biz, derailed 16,000 Teams accounts

ericsmith881

I suspect the issue was more along the lines of him using credentials other than his own, something like a shared admin account or a service account. Such stuff is depressingly common even in big organizations that know better. It's difficult to determine who might know those credentials and a pain to change them if they affect lots of users/services. That's one reason (out of many) why they're a bad idea.

ericsmith881

Re: What an idiot

This is the proper grown-up response. I had an acrimonious parting with two prior companies. In both cases, they failed to change all the passwords or remove access to all platforms. I found out later because I continued to receive alerts (my personal email was on the list for emergencies). When I figured out I could still get into things, I notified them. I even went so far as to work with the outsourced IT firm that had "replaced" me and my team because they were too incompetent to do it themselves.

Yeah, i could have given them the middle finger or pulled what this guy did. I knew enough about how to circumvent detection that I'd have gotten away with it. But ...I'm a professional. There were hundreds or thousands of people at the company I could've caused damage to by acting unprofessionally, and I had no beef with them. Throwing a tantrum is what children do, not adults in a respected profession.

A bridge too far: Passengers on Sydney's new ferries would get 'their heads knocked off' on upper deck, say politicos

ericsmith881

Re: They could

I think that idea is *starting to sink in!*

ericsmith881

Re: They could

You neglected the fact that when the two holes meet, both parties will be upside down with respect to one another. Any Aussie walking out the British side would fall up into the sky. Likewise for any Brit exiting the Aussie side. You canna change the laws of physics! ;)

NASA to stop using names like 'Eskimo Nebula' and 're-examine' what it calls cosmic objects

ericsmith881

You're normal-skinned. Everyone at NASA who thought this was a great idea was either (a) irretrievably thin-skinned or (b) hyper-concerned about appearing woke enough.

ericsmith881

Re: ... and how about the black holes?

...and even if it hasn't been perjoratively used but someone, somewhere *thinks* it may have been - or might someday be -- used perjoratively against anyone, anywhere, regardless of whether they do exist, have existed, or may ever exist.

This is just ludicrous idiocy.

ericsmith881

Who's offended?

The question is, who was beating down NASA's door saying their life was being ruined by the racism/sexism/whatever-ism inherent in the names of *astronomical objects*???

The answer is: nobody. Or at least nobody that isn't a serial offendee of everything, everywhere, all the time. This is virtue signalling, nothing more. And it's just as pointless and stupid as all virtue signalling. It's NASA's stab at "hey, we're woke! Come look at how woke we are!"

What's next? Brown dwarfs may offend height-challenged people *and* people of color! Blue giant is offensive to plus-sized people and those suffering from argyria! Oh the humanity! Won't someone please think of these tortured souls whose lives have been ruined but such vicious, evil slurs??? I'm sure if you try hard enough, someone somewhere will be offended by "globular cluster" or some other innocuous term *if they really want to be offended*.

These people should not be celebrated for claiming to be victims of things that are clearly not offensive. We should not change the definition of "offensive" to mean "anything that anyone, anywhere says is offensive." It's idiotic and it needs to stop.