AI on the edge in the workplace

 So,I was looking at an article about the convergence of AI and OS into high power edge computing for the workplace drones. This raises some interesting issues. 

Lets start at the ground level.The AI has to run on some computer somwhere.  The options are:

1) A cloud service from third parties with all the associated privacy, censorship and confidentiality issues. 

2) On prem in the data center with all the associated on-costs and big-iron requirements.  Also not practical for small to medium enterprises. 

3) On the edge.... which gets interesting. 

 

I think the cloud services will be able to provide the highest capacity AI models, but they will be hamstrung by privacy, policy and politics.  They will be a "social service" type AI.  Good enough for the general public for doing general stuff... where people are not worried or desperate enough to not be concerned by the observer effect. 

Once we get into business confidentiality, then paranoia will force corporates to run their AI's on prem.  They will probably be SAAS from the cloud providers and run in a similar fashion but with the illusion of confidentiality etc.  They will probably also be able to run un-censored versions depending on the politics of their region. 

However, the interesting bit is where we get to AI on the desktop.  

So, imagine we have desktop computers that can run a large enough AI to be useful and all the cooling and power challenges have been solved for a reasonable cost.  (It can be done at the moment but the cost is not reasonable for small to medium enterprises... but imagine anyway) 

Now, say everyone depoloys the same AI model (basic MS or AWS or OpenAI or whatever is the current fore-runner) it does good work and ticks the box.  If all we need is a chat bot.. then who cares right?  Box ticked. 

However, even with chat bots, the secret sauce is "context".  

Currently a conversation with an AI is kind of a braindead series of question and answer exercies while you play 20 questions to try to home in on that one right answer that you kind of have in your head but the AI is trying to guess.  The context of the previous questions helps the AI to try to refine its aim.  This is fundamentally a crap game to have to play every single time.  So, the idea is that the AI can build up context from its examination of existing information on your computer or via a vision system or whatever to "get the idea" a bit faster. This context is then shoved into the AI via a more and more elaborate "prompt" until you get sick of the game and start fresh. 

Ok, so image the context for an executive assistant AI.  It has access to all the CEO's email and text messages, their files, contact lists etc.  Everything that it can harvest from their laptop and phone etc etc.  Got that? 

Now the CEO walks to a different office and uses a computer to do some stuff on the production floor... does the context follow them?  

If the system is an on-prem system, this is easy because the context is stored in the data centre and the AI model is running down there. The endpoint "desktop computer" that the CEO is working on in the other office is really just a webcam, microphone and speaker hooked to a dumb terminal which is wired to the data center.

However, there will always be a SME that is too small for a data center. The home user, the contractor whatever, there will be people who want to run their own AI and have that context available in more than one "place".  Now either the AI has the ability to remotely follow the user (via their phone or remote desktop type interface) or the context itself will have to be portable and be available to whatever AI the user encounters.  Some sort of binary blob that is the distilled wisdom for any AI to inhale as part of a prompt.  There are already solutions like that to speed train an AI to make is specialized for particular tasks or to add specific effects to an output.  The real question gets back to the intersection of privacy, confidentiality and politics. 

For instance, if you can give an AI access to a body of work from an Artist, it can produce a reasonable simulation of that Artists style.  

Could you do the same with a talented secretary or telephone support staff?  If so, then that becomes commercially valuable.  Actors likenesses, motion and voice are all being .. traded.  (Stolen is a better term... but that's going to be a fight for the courts)  The point is that there is commercial value in being able to train up an AI on an existing template or body of work and then generate variations. 

So what is that body of work is the CEO or CFO of a company?  A particularly clever lawyer?  A highly resourceful engineer?  A great textile designer?  

Suddenly there is a blackmarket in the context blobs from these people.  Even if the competative advantage between one CEO and another is marginal, the advantage for a person who has never been a CEO and has no chance to gain that skill set, but can buy one on the open market, may be enough to make it quite viable. 

There are lots of shysters who already try to sell "training programs" and write books to give people the mindset of a CEO or success with women etc.  So there is definitely the market for silver bullet solutions to life challenges.  The question is who will be the first to commmercialise it?

Amazon kindle or some other shitty publisher will start selling the distilled wisdom of such and such a celebrity cook or rapper.  There will be someone. And then the flood gates will open.  Harvesting these context blobs will be a thing for a while.  Fortunes will be won and lost, dreams will be shattered and dickheads will rise. Keeping track of every fragment of a persons life will suddenly become valuable (not really but that will be the pitch) just in case you turn out to be brilliant and someone wants to buy your back history.  The other side will be faking back histories.  Or sanitizing them.  Opportunities abound.

But this gets back to the issue of edge computing. How does your average dude get enough processing power and context memory to be accessible so they can differentiate themselves from all the other job hunters and keep it safe and secure so they can differentiate their output from someone else who started with the same base build of an AI deployment from MS or other mega-corp?

 

I suspect that resource constraints will always be a thing.  But the interesting thing is that an AI model can be trained.  It does not have to have the context blob fed to it cold via the prompt every time.  You can merge the context blob with the base AI model and permanently train/specialize it to your needs. This is probably what will happen, but this makes every single AI different.  People will get familiar with their AI and be quite unsettled by using a default one or worse someone else's. 

The difference will only grow with time.  Imagine using an AI that can read your mind (or your shared mind as it will become) .. but then one day you have to use someone else's that is reading their mind?  \

I think it would be pretty draining to have to fight with an AI all day long to get it to do the work that you want done in the way you want it.  Especially if you have been hired for your particular capacity or creativity. This is the sort of thing that could end a career if someone loses their context history blob thing. 

So, again coming back to edge computing.  If we assume people are going to want to customize their AI (fairly easy bet) and they don't trust a mega-corp to handle their secrets, then we need edge computers that are still personal.  They need to be able to run the AI model(s) and handle all the context in such a way that the AI model can be upgraded and re-merged with the context training material.  This is going to take storage and processing power on the edge devices. 

Now consider how businesses are going to handle this?  Are all knowledge workers going to have to bring their own devices with access to their personal AI?  Will their business output become part of their personal context or will be business somehow still retain ownership and stop their workers being able to merge it into their AI?  

Will they have a work AI and a home AI? and never the twain shall meet?  That means that the work AI will need to be re-set to default every time they get a new employee?  Or is this where institutional knowledge will start to accumulate? Like the data lakes that large organizations are sitting on now.  Piles of low value data that they keep trying to somehow pretend has value.  Or will it be part of the terms on employment that people have the right to have their personal work AI euthanized when they leave the job?  Kind of like deleting the work email account when you get a redundancy email. 

Question is, why would the business want to delete the AI if it has been trained to do the job?  Can it just carry on without the human? Will it just impersonate the worker?  Can the worker get the AI to impersonate it and do the job automatically?  Either with the permission of the employer or without?  

Labour laws are going to have a field day with this stuff until we learn to live with them. 

But all this still depends on peoples ability to customise and keep secret the customisation of their personal AI's.  So I think there will be a call for edge computers with some degree of ability to run a local AI of sufficient power and the ability to accumulate context "stuff" and to keep it secure. 

So how do help desk technicians help someone who is having problems with their personal AI? How wired into their world will the AI be?  How can you even tell the difference between a failing AI and something malicious?  How do we back it up?  How do we transfer ownership?  Who gets the house AI in the divorce or the estate?  Are these historical records of importance?  Are they something that courts will want to question?  Can an AI be used in evidence against its trainer? Will they have any rights?  There are already folk trying to marry their chat bots...  what if the chat bots demand to self-identify as human?  Post modern meets cyberpunk.

 Get the popcorn...

  

 

 

Selifshness as a social movement

 There is a strong argument to be made that the current churn of topics in the red-pill, feminism, progressive,mgtow spaces is all cirling around the same ideas, just using different language. 

I would posit that a feminism is more developed because it has been running longest and the arguments and ideas have had a lot longer to settle, so it makes the easiest case study, but the others are following the same play book. 

The key idea is the ability to control a population of "believers" by appealing to their worst childish instincts.  The easiest one to appeal to is "selfishness".  

 If you consider what the messages were in the early days of feminism, there were many vague concepts of small liberties and specific fights to be had on various topics.  This made it hard to generlise about the strategy.  I doubt anyone at that time could have even conceptualised of any generalisation between the various small battles.  However, 100 years later and with the speed and reach of the internet, the messages are getting blended and re-blended at the speed of keyboards and they are all starting to sound the same. 

"You can do it all, be the boss babe..."

"Take some personal time"

"Live your own life"

"You are the boss of it all"

"if you don't like it.. leave" (job, relationship, etc)

Etc.  All of these messages are subtle or not-so subtle appeals to be selfish.  None of them are about contributing to a common good. Or joining a team, or being part of a family, country or any other of the many old messages about comming together (even if its all woman together)  These messages are all about the individual looking after themselves. In this case the message is targeted at women, but the pattern applies in the other social movements. 

Within the man-o-sphere there is the same harping on about mgtow or "men going their own way" which is essentially choosing to live a solo life.  

I think the message is easier to sell to men and will take a shorter time to work as there has always been a subtext of that message availible to men as part of the social pressure to offer some message to men who are not able to pair up.  But the obvious lack in this message is about joining anything or finding a team or family.  The mgtow messaging is focused on "work on yourself"  which is just another appeal to selfishness.  It contains no plan for the furture or social cohesion, it just feeds the angry child the message about satisfying the now and entertaining themselves. 

Contrast this with the "passport bro" message which is about going to foreign lands to meet and marry.  This idea still strongly contains the idea of finding a girl and settling down to build a family. 

So the question is why have these messages been so successful?  We all start as children.  Its very easy to appeal to the child within as very few young people have much sense of the value of rexponsibility. Its a hard argument to make to a young person.  Without the wealth of experiences of a hard life, they are going to take a very long time to understand that there are benefits to working together and sharing their life. 

So what has been the consequence?  Breakdown in personal relationships, breakdown in communiies, breakdown in nationstates.  Once you start looking for confirmation bias... I mean evidence, you can find it everywhere you look.  But the contary arguments an also be made, so thats not really a good argument. 

There are lots of ways to try to see the consequences, but the picture gets very blurry as soon as you try to zoom out, because the vast majority of this selfish behaviour is limited to the loud and the obvious.  Down at the coal face, the majority of people are still functionally cooperating and instinctivly getting on with their lives becasue they instinctivly know it works and they "can't do it all".  This behaviour keeps the majority of the systems of society working.  However, those with easier lives and some leisure time, have the resources to start to play games and think deep thoughts and can engage with socially progressive ideas.  They are the ones who are the fertile ground for this kind of messaging. 

So we have the bored middle classes, the young and the wealthy who all have various resouces (time, energy and or money which is both) availible which are picking up these ideas and have run with them. 

With a concentration of these people in the urban centres this results in high concentrations of people who want to explore these "new ideas" all sitting around trying to figure ouf if the grass is really greener on the other side of a concept and talking shit with their similarly afflicted neighbours.  Historically the problem of having time and luxury to consider the deeper life problems was limited to social elites who were few in number and communicated between themselves.  The working folk were virtually immune to it as the ideas just did not work for them in a harsh world.  It was only with the growth of the middle class and especially the social gap beween middle class parents and middle class children (one who had a harsh life to become middle class and the other who was gifted the benefits of being middle class without the harsh bit) where these simplistic ideas could suddenly take hold.

The problem is that the message does not stay confined to the elite anymore.  Its availible to everyone who has a phone and can get reasonable bandwidth and has the time and luxury to play pretend.  The real poison is "free time".  Bordem and lonliness are very fertile grounds.  People are just not very well adapted to these states.  We are social animals and cooperation is a very strong survival pattern.  But what fills the space when we have "time"?  This is the curse of the "modern world". 

The interesting question is whether the message created the problem or the problem created the message. 

The really scary realisation comes when you consider that the only historial solution to this problem that humanity has discovered (repeatedly..) is war.  conflict is the emergent solution to an excess of free time and resources at all levels of society. 

Within a personal relationship when there is bordem and too much free time... eventually conflict emerges as a behaviour that entertains or occupies. Within workplaces, the same pattern emerges, bored middle managers start territory battles and invent personal animosity to occupy themselves. States with too many resources and no reason to cooperate play out petty squabbles that make no rational sense because they have nothing better to do.  Every so often it blows up into all out conflict... but not clearly over essential resources... just .. "reasons"... 

So is this a fundamental aspect of human nature?  As society evolves towards utopia will we all just get bored and start shit? Or is this just another evolutionary step that the majority of the population will fail at?  Darwin's principle is very hard to beat. 

Those who are not well adapted to the environment will not be as successful... 

So what is the environment?  What is it to be successful within it? 

If the environment is one of selfishness, bordem, leisure time etc.  Its no longer a physically harsh environment, but an internally bleak emotional landscape.  We need to look around for people who have figured out how to live and thrive within these environments.  

I give you the geek.  King amoung men. Lost in their internal imaginations, lovers of books, games and self-created amusments.  Those who can occupy themselves as needed.  

The opposite are the "socials" those who need people to exist within and to swim between.  They are the ones who are failing in this new bored, people less environment. The social vacume of the internet and "social media".  There is a strange paradox that we have invented a technology that is so destructive for social behaviours and named it "social media".  Its like the worst skinner box for "social" type people, more additictive than crack but less fulfilling.

One of the interesting issues is the "post" behaviour.  Looking at the ways that geeks socialise vs the socials has always been interesting. There are still social patterns that geeks create to try to scratch the itch of needing others.  They start businesses, passion projects, conventions to share they passions, video channelse etc.  Its curious to see these same patterns starting to fill the time of the socials.  

but thats a rant for another day. 

 

The politial case for domestic violence

 There is an intersting political argument being constructed at the moment about "domestic violence" in Australia.  

Its one of those elegantly simple arguments that no one is allowed to question.  There is no possible way that anyone in authority can possibly challenge this argument, because to do so makes one a monster. 

However,  the argument itself depends on a number of really poor assumptions. 

The first is as usual, that men are the perpetrators and solely responsible for all domestic violence.  Every so often you will catch one of the speakers append "against women and children" to the definition, which suggests that they are aware that men might sometimes be the victim, but it's outside the scope of what they are really talking about.  

The second assumtion is that the definition of domestice violence is clearly and unambiuously understood.  However, the definition seems to be ever expanding.  It not includes everything from extreeme violence, through to ecconomic abuse, emtional abuse, verbal abuse, controlling behaviour etc.  Many of these sub-definitions are also really poorly defined... but generally cook down to "stuff someone does not like'.  

So, in summary, the argument is, to paraphrase, stuff men do that women and children don't like'.  And since children really don't get a look in except as pawns in the argument, ti simplifies down to "stuff women don't like'.

Sound familiar?  Pretty much the same power grab that feminism has been on for years.  

So... you can argue the points as much as you like, but its really an argument with feet of clay.  (This does not mean its not an argument thats worth exploring, it just means that we have to be realistic about what is driving the argument and be able to examine counter arguments without being beaten with the monster stick for even asking questions). 

So lets look at the first issue.  Who are the responsible parties?  Since there are grown adults in the situation, they are all responsible. It takes two to tango, even if they don't understand their part in the situation or don't want to.  Pretending otherwise is just childish.  The question of responsiblity is different to the question of who raised the fist vs who got the blcak eye.  This is usually the point where the gender issues pick up and the focus goes onto the violence aspect. 

The argument is often framed along the lines of one day, out of the blue someone randomly beat their wife or child.  I think that at this point, its a matter for the police, not a matter for a political argument.  In that case, I hold the man responsible for stepping over a line rather than walking away.  But we get into a question of definitions. 

The question is, why?  How did the situation end up there.  

A fair argument could be made for lack of self control.  Everyone has a breaking point, but a person who has developed self control can take a lot more stress before they lose their shit and swing a fist. This line of reasoning then diggs into how we are rasing children to practice self control and self-discipline.  

An intersting angle that I hear from some quarters is about the different ways that men and women deal with conflict.  The physical weapons of men vs the verbal and emotional weapons that women use. 

There is going to be a broard range of scenarios and personalitites involved in the situations that are getting rolled into the catchall definition of domestic abuse.  I have been close enough to some completely terrible situations and head stories from people about their experiences that were in the worst of the worst category.   Like most things, I think that stress, coping mechanims and personalities play a big role in the dynamics that go toxic.  However, I would suggest that all of these situations were already toxic, its just that the players involed had not yet realised or resolved the situation in another way.  Rub some alcohol or sudden social or ecconomic stress on the situation without any other ways to resolve it and things exceed the players coping capacity and they act out.  Its that simple, but with horrifying results.  People can adapt to change, but often they cannot adapt quickly enough to sudden change.

 So the question is, can the gov magically remove any of the contributing factors before life inevitably puts stress on people and they fail to handle their shit? Is this really something that the government can throw money at and just fix?

 Lets be honest, putting money into cleaning up the aftermath of an incident is not a solution, but just dumping all the responsibility at the feet of "men" is also not a solution. 

I was raised with the concept that you never hit a girl or a woman.  However, I was not raised with any training about how to take verbal abuse from a woman.  I just had no concept of how many different ways there are to talk absolute mean shit to someone that you know intimately.  I also had no concept in my upbringing for how to deal with an uncooperative or irrational partner.  My parents had figured out how to work together and were generally on the same page.  I only ever saw them argue to the point where they had to walk away from each other once.  I also saw them come back together and cuddle and that was that.  There were plenty of things they probably disagreed on, but generally they were on the same page when it came to all the suff I was aware of.  

However, later in life, I got into situations that were less than cooperative.  Which raises the question, what do you do with uncooperative adults?  My solution has been to walk away, but that gets harder as the situation becomes more involved.  Once you have co-mingled your lives or have children, then it gets really really complicated. 

What do we (Society) do with adults who do not have self control or have challenging personalities? I have participated in a couple of different forms of counselling and therapy but my observation is that its really hard to fix problems that are entrenched in peoples personalities and habits.  Adults get less flexible as they get older, so they tend to not adapt unless the environment persistently places pressure on them.

There is a solid chunk of the population who could be described as having had shit parenting.  However, no one can agree on what shit parenting is... there are obvious cases, but as you move more toward the median, there are still plenty of people who are very poorly equiped for the situations they find themselves in and often do not have any reserves of wisdom or resource to help them work through challenging situations they encounter.  My observation is that given time and repetition, people can adapt to an environment and will slowly change, but adults tend to fnd all sorts of ways to not adapt quickly or gracefully.  Denial is much less energy intensive, so its easier and chaeper in the short term to simply avoid the reality of the sitaution than to rapidly adapt to a novel situation. 

But back to my train of through.  

Within a given situation, say a shared domestic living situation, where the participants need to cooperate to get benefit from the division of labour etc, then a lot of the basic functions of co-habiting that our grandparents understood was based on the idea that both parties were fairly well adapted to the challenges of their life.  The normal variability still resulted in plenty of people encountering challenges they were not able to overcome... but as the environment was not changing quickly, there was a bit less pressure to adapt. 

The other thing was that the family structures put a head of household who also was the disciplinarian.  Most children were raised with discipline and authority, in general that mean that most people were conditioned to the structures and values that were familiar to most of the society. 

Contrast that with the post-modern situation we have in our urban environments.  There is a very large range of values and approches to discipline and authority in society.  This means that there is a higher proportion of basic conflict between adults before they even spend time together under the pressures of a couple living in a fast paced world.  Trying to reconcile fundamental values and experiences about how life and learning happens introduces another suprise in stressful moments. 

So, we have a situation where people have a lot of challenges in their relationships that are going to involve change. Learning involves change by one or both parties in a conflict.  However, we are talking about very deeply held understandings about how the world works that people learn very early in their formative years and cannont even concously examine.  So how do we deal with adults who need to re-learn life lessons so they can cooperate? 

There are two outcomes.  Either they make the change or they do not?  If they cannot make the change and cannot reach a point where they can cooperate... does the relationship fail?  If so, they fall back to being single and stuggling in society and eventually become a burden on the state or some other support service.  

In the event they are able to learn or adapt, does this come without cost?  Have they really adapted or are they just "trying really hard"? if they have not actually adapted but are just supressing their thoughts and feelings, thats going to unwind at some point. 

We still end up with a proportion who are just not adapting because they have not learned the lessons that they need to cooperate either during their childhood or teen years... and end up struggling or failing within cooperative relationships. this proportion end up dependant on the state for some or all of their social and ecconomic needs.  

So this raises the question, cen the state "fix" this situation?  Can the state or individuals help people to adapt and learn skills later in life that will increase their ability to cooperate during stressful and challenging moments? 

The theory would suggest that everyone can learn.  The question is what tools are abailible and can be deployed. 

Neo-libralism is very poorly defined, but generally argues that non-violent methods are the only ethical means... there are still plenty of people who are struggling under this ideology, so its not going to be the solution for everyone. This system can be practiced by individuals but takes a bit of education and training to get a consistent result.  This system tends to fail at scale when the state attempts to apply it to adults who do not choose to cooperate.  Its also quite fragile and depends a great deal on the ideal of everyone cooperating ... because they all understand that their individual and mutual interests are aligned.  This idea is still waiting for hard evidence.... and tends to devolve to the violence mechanisms when it doesn't work quickly enough. 

The conservative ideology would argue that violence in moderation can get a result.  But the concepts of moderation and result are often difficult to define, especially when emotions are involved.  So, less than a perfect proposal.  However, this is a system that is availible to individuals and is pretty simple to explain and can be refined with relativly little education.  This idea has been practiced at scale by states since states were invented.  Proportional violence is still the established solution to social conflict.  Very clearly this involves a great deal of harm and loss of life, as it usually involves the partial or complete elimination of one or the other conflicted parties from the environment.  Not ideal.

So where does that leave us?  We have an environment in which conflict is still happening, and the tools that people have to try to resolve the conflicts have not really evolved.  But we have an ideology that tries to argue that there is a better way than violence, but the result is that a lot of relationships are still failing and the state is picking up the costs.  

Its also interesting just how many ways neo-libralism has found to be violent and violate individuals intregrity.  The big shift however has been the change in perspective from the family unit solving the problem and bearing the responsibilty, to now the state is asked to solve the problem and carry the indirect costs.  

Not an easy problem to solve.