The US military may soon be deciding what is considered fake news on social media. The Defense Advanced Research Project Agency (DARPA), an agency of the US Department of Defense, announced the development of a new program to combat “disinformation” on social media.
The Semantic Forensics program will “develop technologies to automatically detect, attribute, and characterize falsified, multi-modal media assets (e.g., text, audio, image, video) to defend against large-scale, automated disinformation attacks.”
DARPA is looking to develop software that can detect fakes amongst 500,000 stories, photos, videos and audio clips. The plan is to undergo four years of trials until the software is ready, so the program will not be used in the run-up to the 2020 election.
Only time will tell how accepting social media sites like Facebook, Twitter and YouTube will be of the DoD determining what is considered fake news.
The program is likely a reaction to the claim by Robert Mueller that the Russian company, the Internet Research Agency (IRA), led a social media campaign with the aim of “sowing discord” to help President Trump get elected. Although Mueller claimed the IRA was taking orders from the Kremlin, he provided no real evidence to link the IRA to the Russian government.
The actual influence of the IRA is much smaller than what Mueller alludes to in his 2018 indictment of the company. According to a Senate commissioned report by the firm New Knowledge, just “11 percent” of the IRA’s total content was election related. The IRA’s posts were “minimally about the candidates” with “roughly 6% of tweets, 18% of Instagram posts, and 7% of Facebook posts” having “mentioned Trump or Clinton by name.”
And, the gov. will decide what is and isn’t disinformation. “1984” has become the handbook for US domestic policy and trashing the US Constitution in the process.
Wouldn’t want election based information to just be plainly sourced, now would we ?
Sounds like an excellent rating system for radical journalism. If the Army says it’s “Fake” you know its good. I hope I make the cut.
Oh I’m ssssuuuurrrree you can the Pentagon’s word at what’s true and what isn’t.
The US fighting ‘disinformation’? REALLY? How interesting it will be to find our own agencies locked in mortal combat over who is telling the truth.
Like putting the fox in charge of the henhouse.
I get it… The Pentagon is going to now find a way to convince everyone that their new F35 strike fighter is not really the dog all of the testing has shown it to be…
How does it make sense in the light of Constitution, where Press is the institution mentioned by name as means if insuring democracy. Once Government gives itself the right to legislate and administer arbiters of truth, the whole system is not sound any more.
And after report was written, the Russian internet company in question was in fact indicted. But they sent lawyers to turn themselves in — and it was refused. Apparently, just because you are indicted you cannot turn yourself in — unless indictment was SERVED. The judge listening to this explanation accepted Justice argument, but Justice was forbidden to use the arguments in the indictment again against he company. This is why Mueller avoided it his testimony.
However, the judges decision apparently is not binding to third parties to keep on talking about this “influence campaign” as if evidence based, and adjudicated. If you cannot even turn yourself in — to get your day in court to face your accuser, how do you defend yourself? Or worse, how do you defend yourself against algorithm?
Spot on. And the algorithm is rather simple: indicted = guilty. Unfortunately for the indicted, it is well known in mathematical logic (and so in law) that statements like this cannot be either proved or disproved by logic itself. Hence courts. But then, they use another algoritm(s). And so ad infinitum, while the human life is finite. Hence, as you stated, the algorithm, i.e. its owners, always win.
Indeed, owners win. Technological advances, as well as the sense of invulnerability, creates environment of highly centralized power in very few hands. But as any Hubris, it generates its own Nemesis. And the Nemesis comes in the shape of Ignorance and Want. There is no fertile soil to nurture science and innovation. Owners want quick profit, slamming their fists at the table demanding their serfs innovate. And serfs just want a job that pays well and is well defined. And owners do not understand. What is wrong with people, so they want to replace them with some other people, more entrepreneurial . It will never occur to them that it is their fault. That the very system they created was tilted to make them successful — the producers of everything useful on earth! The very Ayn Rand type of pure success! How can they be at fault? And as the unworthy start complaining — it is time to put an end to this. Lest people figure out they are led by people who — in spite of the wealth, are incompetent.
Pozdrav,
Bianca
PS. Regarding the article itself, no comment. People either have never heard of Goedel’s Theorem, or intend for the AI to use their database of true news (pattern recognition). So, again, an Einstein would have been pronounced fake, or there would’ve been thousands of them every day, depending on the algorthm.
Hvala Bianca. I agree with you very much about your views on the link between fundamental and applied (i.e. technology} research. The “bosses” want technology, but regard fundamental research as a waste of money (although I suspect there is more to it than just money). I have fought it (and suffered) all the way since Thatcher’s (3rd class in chemistry from Oxford) dismissive coinage of the “blue sky research” in ‘ivory towers” phrases in the eighties, seeing with my own eyes the results of this policy. There are very few Physics departments left in the UK, and mathematics graduates can only find jobs in banks. Very few domestic students continue to the PhD level, and are replaced by foreign students to supervisers who are too busy to look after them, and being interested primarily in increasing the numbers of “publications”, not their true quality. The rules require that they finish in three years, and the quality is consumerate with this straight jacket (“market discipline”) requirement. (As for quality, there is a bureaucratic algorithm – the number of citations, but because the whole thing is a cartel, it is done by “you scratch my back and I scratch yours”, so the poor Einstein would’ve never been heard of). Similar thing followed in the US since the nineties. The “leaders” are incompetent because they inherited everything, not having passed through a stage of their own struggle to create something. I have been on both sides of the coin, so I know first hand. Anyway, I’ll continue enjoying to read your comments on this and other sites you post them. Puno pozdrava, i jos jednom hvala, Nebojsa.