Also of interest to @GCHQ, @csiscanada @AIVD @CyberGovAU.
cc: @ericgarland
With apologies for the length of the following, and the unpolished prose (it's the second of four sent over four days, and not remotely the longest)...
Here it is...
Falling Behind: How Social Media Companies Are Failing to Combat Inauthentic Behaviour Online
stratcomcoe.org/how-social-med…
port presents a different perspective: We were easily able to buy more than 54 000 inauthentic social media interactions with little or no resistance.”
…
vast majority of the experiment by buying engagement on inauthentic profiles we created ourselves.
Be sure to assemble a strong legal foundation for these efforts, but there are undoubtedly enough reasons to be investigating many if not all of these operations.
/End Quoted Message Transmitted to the FBI
As no one has asked me to remain silent regarding this or many other messages to the @FBI, it seemed worth sharing.
Obviously, there are ways to supercede this.
And apologies again for the lack of brevity, by Twitter standards, but the above amounted to three pages in the real world.
Think about that.
cc: @RepAdamSchiff @OCCRP @ARCYBER @USTreasury @TheJusticeDept
"Spending 300 EUR, we bought 3 530 comments, 25 750 likes, 20 000 views, and 5 100 followers, enabling us to identify 18 739 accounts being used for social media manipulation."
The FBI has known since at least 1/2/20.
It's *trivial* to find *vast* numbers of accounts being used to supply purchased engagements.
Via both the other engagements bought from known networks.
Via tracking *those* accounts back to active botnets and other sources, and then identifying *other* accounts under their control.
Now realize, as Roger Stone's networks get pulled down...
Companies and individuals may already be proffering out on this issue.
Now how many of those were Russian organized-crime fronts, or Russian-intelligence fronts?
Or both?
Or even the second.
And one of many such messages, most far more significant.
Nevertheless...
Indicating they found several thousand or so commercial actions...
How many thousands of public figures, of varying degrees of prominence?
With a dash of, say, espionage?
Which has *no* statute of limitations?
What a *glorious* way to pull in vast, disparate masses into some form of criminal conspiracy.
Like publicly vouching for your product, your service, or your candidate.
Or your international policies.
Or what have you.
And yes, speaking of self-evident things.
If this is coming out now - even taking into account 6-months of lead time for the @FBI and no request on their part to keep it quiet...
Still, while that alternative is balanced and honed to this purpose, this open vulnerability...
At least now.
And again, there's *so* much more afoot.
Sleep well.
The sleep of the just.
cc: @NatSecLisa @NatashaBertrand @ZevShalev @NSAGov @BarackObama @SpeakerPelosi
If you're involved in RICO conspiracy, you're evidently culpable for all the crimes undertaken by said conspiracy, including even the ones you're not aware of.
Both of which seem critical if you're ever involved in a vast, sprawling RICO case which is elaborately documented.
Speaking of which, while there are private Facebook groups and so forth, all of Twitter up until the end of 2017 was being archived in the Library of Congress.
Hence, it's a bit late to delete your pre-2018 tweets.
Speaking of which, cryptocurrency like bitcoin is a permanent, public, online, unerasable record of every transaction, held forever in the blockchain.
Or if trolls, companies and/or the controllers of botnets were paid via cryptocurrency.
And will cease to be a bank.
Critical if you are tracking bank transfers, but often unnecessary.
Some of this would apply to tools such as laundering payments via clickfraud, though there are other methods.
Combined with server logfiles holding data for up to a year, how many of those slow-running PCs mothballed in the attic hold evidence?
And that's leaving aside the US counterhacking and taking over a botnet, such as North Korea's Joanap.
Beside trying to overthrow elections, undermine public-health measures during a pandemic, and so forth?
There's something evolutionary algorithms in psychological warfare.
But you can also do that with people & the stimulus directed at them.
Where you have extensive data on views, reactions, further shares, comments and so forth.
What can you use this for besides analyzing individuals?
Shaping their mental state.
What happens if you do the same thing with stimulus in psychological warfare?
Aside from incredible harm and evil?
But the computers can literally detail every minute detail & action the felonies, & literally record the whole thing.
Vast, complex PSYOPs managed with the precision of supercomputer.
Imagine if the records of the actions undertaken, and even the computers doing it, could be seized.
Not just the evidence but the crimes themselves, frozen in evidentiary amber.
Think about all the precise pieces of stimulus not only visible as being managed, but easily tracked across a host of servers and databases, as very specific code came up again and again.
Showing further collusion in the evolutionary algorithms, botnet support, bitcoin payments, etc...
Mentioned repeatedly in an SF context in 2011.
I literally shelved a lot of positive applications to avoid this one.
eclipsephase.com/comment/20628#…
People expressly vulnerable to this manipulation.
"Undoubtedly someone will do a mass public search - or *has* done a mass search - on all #cryptocurrency identifiers. It's hard evidence. Other data points will crop up. Emails, names, pseudonyms, organizations, businesses, hang-outs."
As you've broken it *all* down into discrete packages - #bitcoin searches, commercial encryption, PSYOPS stimuli - for them.
But as we animate it across time, it becomes more than a sculpture of light, but a recreation of the living crimes themselves.
But far from frozen.
These are "embarrassingly parallel problems."
Parallel processing through clusters (Beowulf, Hadoop, etc) rarely reaches its full, overwhelming potential.
Problems are rarely broken down into a host of simple parts for them.
Especially not at this scale.
And worse, not only is the evidence online, and in formats easily assimilated and merged...
But they were doing all of this on the cusp of quantum computing, and particularly quantum search.
Grover's algorithm.
Either in the near future.
Or now.
Running social-media psychological warfare w/de facto evolutionary algorithms is simple enough, even if you don’t know what you’re doing.
Existing marketing metadata has already been honed for this work.
Keep switching it out based on quantifiable reactions & intended goals.
Think about how many elements are endlessly repeating in even a minor feint.
Data packets.
Elements revealing in themselves, given who uses them, with what timing, and with what coordination in the outside world.
But how many were transparently created for this work?
And do the records at Cambridge Analytica or elsewhere indicate that a particular clip was pre-planned?
Do they reference it before it happens?
& does *anyone* reference events like mass shootings before they happen?
Including who is involved, and *when.*
What they may know, or their actions & data may reveal, in turn.
Coordinating on this scale and effectively recording the whole thing means you practically have the crime itself, not just the evidence, graven in stone.
The sheer scale of traceable, relentlessly repeating evidence, all of it stored on a host of insecure devices, even if your keystone processors are secure.
But providing your adversaries with an ocean of digitized and analyzable hard data regarding your activities, especially ones thought invisible, offers immense insight into your operations and goals.
Even if they’re taking opposing sides?
Or when they’re driving hacks or DDoS attacks against specific industries?
Not by themselves, but superimposed against that larger set of databases referenced above.
#Botnets would show distinct changes in the operation of individual computers but also changes at scale in masses of them.
Always on, strange operating hours, changes in sites accessed.
& determine responses to counter, disrupt and deactivate them.
Having numerous botnets to reference would tell you more.
What if you have no evidence or testimony to proffer on it that law enforcement needs?
Someone will talk.
Or their emails will.
What about the people who are left?
If they dig really deep?
So many of these ecosystems - DC, LA, NY - conspiracies and criminal operations may seem small and tightly interconnected.
With evidence on *everybody,* tipping points hit when *everyone* knows they should proffer.
Things could get very interesting, as proffers and evidence unearth completely unrelated crimes.
But they are significant and far reaching.
More will emerge into the public light, in good time.