4 ms·

I call bullshit on the "potentially cost 80 people their jobs". For any "old-school insurance company", I'm confident there's a hell of a lot higher standard f

by powera 11y ago

I call bullshit on the "potentially cost 80 people their jobs". For any "old-school insurance company", I'm confident there's a hell of a lot higher standard for firing 80 people than some program "at least as effective as our analyst team in a majority of cases" according to some IT guy.

9935c101ab17a66 11y ago
It may have been slightly hyperbolic, but in your vehement over-reaction, you literally quote the word "potentially," so maybe you should simmer down.
vonklaus 11y ago
> I'm confident there's a hell of a lot higher standard for firing 80 people than some program From where do you manufacture this confidence? Insurance is a highly competitive business and we/you know virtually nothing about this company. No financials. No growth metrics. An unknown number of employees. Leverage? No. Cash on hand? nah. We have a few sentences from a narrator with less credibility than Holden Caufield throwing out some lofty claim that sounds a lot more like a survey disguised as a question, than an engineer wrestling with the magnitude of a breakthrough. I would hesitate before projecting confidence with this little information. There are many hypothetical situations I could imagine that run counter to your appraisal, but I could never get comfortable with them because with no data or info it would be quite baseless. Caveat, I believe the main post to be utter bullshit. I would assume someone thoughtful enough to build this system would likely have considered this before working for several months on this without consulting anyone in the company. Further, s/he doesn't participate in the discussion at all which is what I would expect from someone wrestling with an ethical issue. They would have likely made a decision and tried to justify and defend it in the comments section. Sounds like they just whipped up a GUI using Visual Basic and used the IP address to generate the TPS reports using real-time-big-data-pattern-matching-artificial-intelligience-NLP.

by powera 11y ago

9935c101ab17a66 11y ago

It may have been slightly hyperbolic, but in your vehement over-reaction, you literally quote the word "potentially," so maybe you should simmer down.

vonklaus 11y ago

> I'm confident there's a hell of a lot higher standard for firing 80 people than some program From where do you manufacture this confidence? Insurance is a highly competitive business and we/you know virtually nothing about this company. No financials. No growth metrics. An unknown number of employees. Leverage? No. Cash on hand? nah. We have a few sentences from a narrator with less credibility than Holden Caufield throwing out some lofty claim that sounds a lot more like a survey disguised as a question, than an engineer wrestling with the magnitude of a breakthrough. I would hesitate before projecting confidence with this little information. There are many hypothetical situations I could imagine that run counter to your appraisal, but I could never get comfortable with them because with no data or info it would be quite baseless. Caveat, I believe the main post to be utter bullshit. I would assume someone thoughtful enough to build this system would likely have considered this before working for several months on this without consulting anyone in the company. Further, s/he doesn't participate in the discussion at all which is what I would expect from someone wrestling with an ethical issue. They would have likely made a decision and tried to justify and defend it in the comments section. Sounds like they just whipped up a GUI using Visual Basic and used the IP address to generate the TPS reports using real-time-big-data-pattern-matching-artificial-intelligience-NLP.