Image: Computer code (cropped) by rawpixel.com
Who controls the data the government collected from you for a generation?
Your insurance company collected data on your driving – so did your Lexus – who owns that data?
You told your doctor about controlled substances you used – and now it gets brought up in an interview.
If you can’t exclude someone from using your data, then you don’t control it. That means you really don’t own it. It’s that simple.
What does “own” mean here, let’s define the terms.
Owning the data means you can do anything you want with it – share it, sell it, mine it or build an A.I. language model with it.
From birth until the last Social Security check gets cashed, your data is collected by federal and state agencies, corporations and of course the internet.
Your teen daughter puts every waking moment on Facebook or Instagram – so who owns those hundreds of images?
TSA Pre Check, Medicare/Medicaid, Social Security, government or military retirement, Tri-Care, veterans hospitals, and of course, the IRS – gather more data about every citizen than has ever been gathered in the history of mankind.
Each agency gathers different data, at different times, for slightly different purposes. And those purposes may change over time.
Who owns the rights to that data?
It’s a far stickier question than you think.
The knee jerk response is the government owns the data. They collected it for their purposes, so it’s theirs.
The government will certainly say so.
What if the “purpose” changes?
Your personal data comes in several forms – that which you expect to be “not private” like your birthdate, address, employment history.
It also comes in that “hey, that’s my private business” kind which you may have to cough up to the insurance company and is picked up by Medicare – you are HIV-positive or you have a heart condition, or you are a recovering alcoholic.
When you provided the information it was for a specific purpose.
A reasonable “need to know” existed.
Now that purpose changed and people who have absolutely no need to know you are fighting alcoholism – gallantly – will see that info and perhaps make adverse decisions with it.
Today artificial intelligence is training models called LLMs – with all that data. It’s not data that they called you and asked you to provide – voluntarily.
Does the government have the ownership rights to your data, collected for innumerable innocent purposes, now aggregated into a database it will use to train an A.I. model to surveil you, to predict your reaction to stimuli?
To assign you a “good citizenship” score?
Go with us here because we at Fractal are right in the middle of this and we see spectacularly scary implications.
Our favorite Vice President is in the middle of it too – as you will see in this Twitter/X video all about Palantir:
https://x.com/ProjectConstitu/status/1985066976020922487
All that cool A.I. stuff in Chat GPT is fun until you realize the government can train massive A.I. systems on decades of data about YOU – and there is nothing you can do about it.
Even worse, you do not know what’s in that data about you – maybe it’s something completely wrong – or a bunch of wrong stuff – and that training model can stop you from getting an FHA mortgage or your kid getting a student loan.
Will Social Security or Medicare payments someday be dependent on good citizenship scores?
In the commercial world, A.I. developers are encountering the concept of “data sovereignty” – something we covered in an earlier post.
Data sovereignty is important because a company or government agency must put its data into a LLM (large language model) to develop an A.I. agent, for instance – and once it’s in that model, it stays there.
Who owns it?
That really means who has access to the data and who can use it.
Not the entity who provided the data for the LLM. It’s gone from their control.
It’s like the 1958 movie The Blob – everything the model ingests makes it bigger and scarier – remember that movie theater scene?
(By IMDB, Fair use, https://en.wikipedia.org/w/index.php?curid=880563)
The Fractal team and some hardware vendors are developing a concept beyond corporate data sovereignty – we are dealing with personal data sovereignty.
You may recall Elon’s D.O.G.E. guys found bad guys operating in several government programs – but they did not have the technology to determine if those bad actors operated in lots of government programs, because the government databases are not integrated.
In some ways, the old information is fire-walled – but that is changing by the minute.
We do have that technology and could solve that challenge easily.
Trump gave Palantir a no-bid $30 million contract to tie those program rolls into One Big Beautiful Database.
So now all the data, collected about you, your family, ancestors for generations – is in One Big Beautiful Database – and a bunch of it may be wrong!
Did anyone from Palantir call and ask your permission?
What do you think they are going to do with that database?
People, the government is building – or can now build – the world’s largest citizen-level A.I. training model – its input is your personal data – dating back to your birth.
They may claim they aren’t going to run training models – but once it’s built, there is no way to stop its use for training A.I.
The government is moving from the surveillance state to the predictive state – knowing in advance how you are likely to react to stimuli – and you coughed up the data and you are getting squat in return.
You don’t control it – so you don’t own it.
META runs training models on Instagram and Facebook data – because they own the data.
Elon runs training models on Twitter/X data – because Elon owns the data.
Everyone knows that but since nobody expects Twitter or Facebook data to be private, people are OK with it.
Palantir may be running training models on you – how’s that feel?
Are you OK with all your health care info being integrated with your tax data and your voting history and your political donations?
Do you control how that data is used?
We at Fractal think new technology provides a less intrusive opportunity and we want to intro a concept we think you should consider as an alternative.
So follow our Substack since we are publishing more on individual data sovereignty.
First a personal note.
The day of public data aggregation is here, by good actors and by bad actors, and you cannot put that genie back into the bottle.
Once social media encouraged millions to post daily routines and travels, and FOIA requests forced government databases into the public square – the outcome became inevitable.
The only question is: how do I control data I gave to the government, and corporations in good faith from being used by A.I. to possibly harm me?
If your reaction is “let’s pass legislation to stop people using our data in LLMs” please stop here, go do something restful, because you are delusional.
There is no way to stop this future – our thesis is there may be a way to manage or control who owns YOUR data.
That is a much better fight.
Technology is at an inflection point and training A.I. systems on data gathered for entirely unrelated purposes tests the ownership limits of that data.
The government needs to know certain stuff about citizens – but using that stuff for training models to predict behavior is beyond the permissible scope of the accumulated data.
But the government isn’t the only one with that data.
That data is in voter rolls, Medicare/Medicaid, DMV and hundreds of other databases many of which are public. Just check out any of the three credit reporting agencies – they have lots of data on you.
We have tons of it – we run the voter rolls for over 2 dozen states, the property tax rolls for major metro areas, the entire Federal Election Commission database – and a bunch of other public data.
We track every NGO, every officer and all connections. You can see videos of that work on the Omega4America.com site now.
Since we handle databases far larger than anything our pals at Palantir or Oracle can handle, we have an educated opinion.
You know from reading this Substack, Fractal is a massively distributed system that enables the largest systems to be moved to an environment where data centers can be eliminated or dramatically reduced.
We have an entire site at TheSustainableComputingInitiative.com with our content on how we do it by impacting I/O wait states.
So, here is a possible alternative to consider.
We can create for every person an individual Fractal – with all his/her/its data in it. We take that data from those legacy government systems that have been busily vacuuming up everything about you.
The scary part is over – here comes the comforting part.
Every citizen’s Fractal reports 100% of the data the government has on him/her or it – that the citizen provided. Obviously stuff the citizen did not provide like law enforcement investigation would not be ingested in the first place.
All we are ingesting is the stuff Trump is paying Palantir to “integrate.” It’s your data provided BY YOU.
You, the citizen can see any incorrect data in a government database – for the first time ever. You also see where you appear in many government databases – for the first time!
Bet there are going to be some surprises.
You own AND CONTROL – the data in your Fractal – all of it.
You can choose to share it with a government aggregation project if you choose or you can choose not to share it at all if it is beyond the scope of a legit government purpose.
It’s not all or nothing.
The power of Fractal goes to the individual data element level. You can share most of your data but that stuff about the alcohol rehab you keep separate – nobody gets to integrate it.
The government may mandate your data participates in some efforts – they do whatever they want – but you know your data is correct, you know it is being used for a specific purpose – none of which you know today.
You have personal data sovereignty. Kind of like Bitcoin for data – but for your personal information – not for crypto.
How do we implement such a crazy thing?
We implement the Great Big Beautiful Database with Fractal technology (at 1/10th the cost of doing it any other way) and when we move the data into a Fractal for every person, we then need a place to run the Biggest Baddest Government Database Ever.
We run it on the world’s largest supercomputer – one without a data center!
We run that database on a MESH – as your Fractal is on your desktop, you phone, and your watch – your Fractal runs on any computer you authorize.
If you are the one guy with no tech, living off the grid with your sheep and chickens, we run it for you.
Everyone has a Fractal – with their data in it – in THEIR control, not the government’s, and certainly not in the control of a software vendor or cloud provider.
If the government wants to run an A.I. training model – it can aggregate the entire MESH of every citizen in America.
However, citizens can “opt in” or not – at the data element level.
The Big Boys in the government may force you to provide your data – that’s beyond our control – but for the first time ever, you know what’s in YOUR data, what the government types are doing with it, and you know your data is correct.
What we want to get across here is the concept of PERSONAL data sovereignty – in the age of A.I.
It is something that can be delivered right now and it has huge advantages over ancient relational databases, in behemoth energy-consuming data centers.
We are engaging you – to start talking about PERSONAL DATA SOVEREIGNTY – because we can do it, we can show it to you, and it is within your grasp.
We here at Fractal and our growing band of partners are encountering more of these outcomes not possible with any current technology – at a fraction of the cost.
In this case, we could close many centralized government data centers and cut billions a year of waste out of government IT budgets.
The U.S. Census – instead of being done once a decade – could be done in real time from everyone’s Fractal on a continuous basis – with far better accuracy than what is done today.
As we post new outcomes – customers are continuing to bring us new application opportunities that are impossible (because of either cost and/or scale) in Palantir, Oracle or other 1980s technologies.
Remember that next time you read the government gave one of “the usual vendors” a $10 billion contract.
We don’t want to own your personal data – we don’t care about the alcoholism thing.
We just want you to know you are on a trajectory to lose all control of your data – for A.I. purposes you never imagined, and maybe you might want to consider doing something about it.
FractalComputing Substack is a newsletter about the journey of taking a massively disruptive technology to market. We envision a book about our journey so each post is a way to capture some fun events.
Subscribe at FractalComputing.Substack.com
Fractal Website: Fractal-Computing.com
Fractal Utility Site: TheFractalUtility.com
Fractal Government Site: TheFractalGovernment.com
Fractal Sustainable Computing: TheSustainableComputinginitiative.com
Jay@FractalWeb.App
@FractalCompute
JayValentine.com
Portions of our revenue are given to animal rescue charities.
The post Palantir, Fractal And Your Personal Data Privacy – Get used to being used, because YOU are the product appeared first on The Gateway Pundit.