It is a combination of words and as the BBC keeps on getting it wrong for all the greedy reasons it is engrossed in, lets recap that setting that they open up with ‘Google staff resign in backlash over its AI contracts with military’ (at https://www.bbc.com/news/articles/cmreyvepnv0po), so they are willing to get in bed with organized crime and hackers (a speculative thought I am entertaining) whilst the military is out of bounds? So as we are given “Google is facing a backlash from some staff members over the company’s deployment of its AI technology for military use.
The BBC has spoken to several tech workers who have either recently resigned from Google’s AI company, Google DeepMind, or are in the process of leaving, over concerns with the tech firm’s military contracts. Sofia Liguori, a London-based research engineer, is one of them. She resigned from Google DeepMind this week and spoke about her concerns exclusively with the BBC, despite her contract restricting her from doing so.”
So lets recap this for the people who have no IT skills and optionally a child with a Nintendo. There is no AI, all AI is fake. There is of course predictive analytics and at present it might be considered Advanced Predictive Analytics, but the difference it that AI and make leaps of faith (according to Alan Turing, not the comedian) and it can do this absent of data. Predictive analytics requires data at all time, in absence of that it becomes Presumption at best and mere speculation at worst. For AI to become real it will the following things.
A quantum computer (we have that)
Shallow circuits (we seemingly have the too, although I reckon it is still in its infancy. IBM had the hard parts figured out in 2017)
A Trinary system and about two decades ago a Dutch physicist created the Epsilon particle, but between that an having a processor/Coprocessor to get it in real life is likely still decades away
So we are not even close to having AI as such I label ALL AI fake. It is not bad will, it merely is and on top of that the validation and verification is far away from all that, as such the ‘guess’ with of these military systems is making it a dangerous setting being anywhere near 25 miles of such a place. As such is this Sofia Liguori really concerned with the setting of these predictive analytics or is she worried about what the client will look for when they are analyzing the logistics of these predictive analytics, because that is merely one step away from what the logic of a fake AI is, towards the analytics of these predictive algorithms. And these defense IT boffins will put two and two together and they will get really different results. That is when they are done with analyzing the verification and validation modules, I found them lacking on several levels last year and I have no faith that they improved anything near a serious level. I tend to blame the lack of a trinary setting. But that requires some explanation. In binary we have Yes/No or True/False. Now in a double setting we get for example the setting of Toronto (Yes/No) and Maple leafs fan (Yes/No) and there would be an abundance of Maple Leafs fans. But that setting is absent of Hockey Fan (Yes/No) and perhaps Recent immigrant from Montreal (Yes/No) win this setting the non-Maple Leafs fans are riddled with inconsistencies for all kind of reasons and whilst you can set the questions right, the model becomes inherently more complex. And this is based on two settings. So to capture it all you will get an exceedingly complex metric of answers. With Trinary it becomes a setting of Yes/No/Both and the equation simplifies. The setting of validation and verification makes it simpler as well, because nuance (to some degree) takes shape. Now in military settings it does not become easier, the setting is more complex and the influence that [BOTH] is rallied towards this and can be counted and as such the tally becomes easier to see that there are additional factors in play, but still the absence of data cannot be countered at present and as such all AI is fake (for now). And the response of ““Realising my company is pushing more and more for AI systems in military operations makes me feel ashamed to work at Google,” Liguori said, speaking in a personal capacity.” Is even more debatable. Because fraud detection is a valid operation for these special forces of the Military Logistics department and they could use the better part of predictive analytics, it is not perfect but they will find optional danger flags really quickly. More work is needed but where 50 people find 12 red flags in a month, One computer will give them 50 red flags in 12 seconds. It doesn’t mean that they are all red flags, but once the setting is checked, it might be an optional misfiled paper, but checkers would have known this and as it stands military logistics could use the help (in nearly every nation) even if it was merely to find 13 billion in misplaced oil revenue. So when we get to “After working for the company for three years, Liguori said she decided to resign over Google’s increasing work with the US and Israeli militaries. “I don’t agree with this direction and this has led me to lose faith in my company being true to its stated ethical values,” she said.” So how did the Israeli military included in that? Was she a supporter of Hamas terrorists? Or merely anti Israeli? I do not know, but the direction of that interview took a wrong turn and the BBC was seemingly OK with that. We see that continuation in “Naza, which examines the role of AI-powered targeting systems to bombard the Gaza Strip, resulting in high civilian casualties. An Israeli government minister has said the claims presented in the film were false and distorted.” So what were these AI-powered targeting systems? Who made them and what are they called? It seems that the BBC has a lot of explaining to do, not merely on this part and to include Google in this is (as I see it) mere fabrication from a delusional mind and I hope that the people will take my explanation for these settings in a proper light, so that they can see that these AI systems are merely fake and there is no real timeline when they will be real and as I see it, such systems will not be active before 2031.
So have a great day and don’t forget to ask your barista if his La Marzocco GB5 S Auto-volumetric 2 Group Commercial Espresso Machine is AI powered.
