Solution:

1. Run a large number of verified unpoisoned images through the AI poisoning algorithm

2. Create a large high-quality data set consisting of the before and after poisoning image pairs

3. Train an AI using this data set so it can detect or even reverse the poisoning

4. AI train stops for no one

It probably can be simplified to writing an algorithm which detects that image was generated by AI (like https://deepmind.google/discover/blog/identifying-ai-generat...), and consider all AI-generated images as bad for training set by default.

That would certainly be the ideal scenario, but the issue is that it doesn't translate well to corporate bureaucracy.

In a sizeable company, your work performance is typically judged based on a small set of quantifiable metrics. How many sales dollars did you generate? How many widgets did you produce this quarter? Did you hit or exceed your set quota for X? and so on. Your work performance is also judged on how well you follow protocol and abide by the company handbook.

Innovation and "hustle" are not quantifiable metrics and get completely lost within the complexity and scope of managing a multi-location cubicle farm. A high level manager who implemented this kind of policy would most likely end up terminating the very kind of people he supposedly wanted to reward and promote.

Pretty much, this policy would only fly in a very small company where a very competent and attentive CEO is personally overlooking everyone's work and gives everyone a very long leash to work with.

In the corporate world, if you try to innovate or attempt to improve on an established process, you're most likely going to face disciplinary action unless you are very high up the ladder. C's and D's (and sometimes E's and F's) are A's, Actual A's and B's are insubordinate and unprofessional.

I dunno man, you're going to get some weird requests on there. Are you ready for that?

>They never stop to think that, if their moral code calls for killing all these people, maybe they should be questioning their moral code.

They certainly do think about it. The logic is invariably:

1. These are people who are holding back progress towards a future in which (according to my ideology) everybody is happy / equal / free / (insert vaguely defined ideal state here)

2. Therefore they must not want people to be happy / equal / free / (insert vaguely defined ideal state here)

3. Only an evil person would not want people to be happy / equal / free / (insert vaguely defined ideal state here)

4. Therefore they are evil

5. Therefore I am morally obligated to annihilate them

6. I am a Good Person because I am so devoted to ensuring that everybody becomes happy / equal / free / (insert vaguely defined ideal state here)

Education won't improve IQ scores. You would just have stupid people that have memorized more things.

Well, it ultimately depends on how you want to define insanity.

Most of the school shooters aren't schizophrenic. They usually aren't complaining about hearing voices or hallucinating. When the media claims they had a "history of mental illness" it usually means mild to moderate depression, anxiety, ADHD, things of that nature.

Some of them will claim to have a political or ideological cause but most of the time their manifesto (if they bother to write one) consists of something like "Everything is meaningless bullshit and I am going to prove it by killing a bunch of random people for no reason".

It depends on who your main source of info is. Right wing media will only cover BLM protests when they turn violent. Left wing media will only cover BLM protests when they are peaceful.

It's so very nice to hear that big tech companies are being so careful over which authoritarian race-obsessed collectivist ideologues their ads appear next to...

So basically the influencers would get paid for doing fan service, while not actually having to interact with said fans. Makes sense that there was so much demand coming from that direction.