2020 election: Artificial Intelligence has chosen a winner - but there's a catch | The Independent
S E E D E D C O N T E N T
AI will one day be able to predict the outcomes of the electoral college with near-total precision: it's not a matter of if, but when, according to statistical physicist Hernan Makse
Artificial intelligence has chosen a winner for the 2020 presidential election — but there's a catch.
Hernan Makse is a statistical physicist at City University of New York who runs the Complex Networks and Data Science Lab at the Levich Institute in Manhattan.
His lab uses AI to predict the outcomes of international elections using social media traffic, focusing mainly on Twitter, a platform with over 48 million monthly active users in the US.
"We usually start one year from the election, and then we use that data to train the machine and predict the outcome of the election at the national level," he said in a recent interview with The Independent, noting how AI can now also be used to predict local and state election outcomes after data is organized by geolocation. "Predicting elections is, of course, quite complicated."
The physicist is the CEO of KCore Analytics, an AI platform which hosts live election predictions and has even begun launching efforts surrounding contact tracing for Covid-19. The company uses over one billion mined tweets for each project it launches, according to the KCore Analytics website, and successfully predicted last year's surprise election outcome in Argentina.
Mr Makse was preparing to release KCore Analytics' final predictions surrounding the 2020 presidential elections in the US when we spoke on Monday.
He revealed former Vice President Joe Biden had a strong advantage in terms of the popular vote, according to the AI models his team had trained, but when it came to the Electoral College, "the situation is different".
Whereas the AI models showed Mr Biden with a healthy lead over President Donald Trump in the national vote, by about eight or nine points, Mr Makse said he "has an advantage, but it's very, very small" in the electoral college.
"In fact, it's so small, that we're actually trying to figure out different scenarios," the physicist said. "It's still an open question for us."
In some ways, AI suffers the same flaws as traditional polling and surveying options surrounding elections and political campaign cycles nationwide.
The same voters who live in rural communities and are otherwise unlikely to be reached by a pollster are the same people who do not often access social media platforms like Twitter, Mr Makse said, effectively excluding them from the data.
"In the case of traditional polls, there are certain groups that are very difficult to capture, and they are the ones who will ultimately decide the fate of the election," he said. "The people in rural areas, first of all they do not pick up the phone when they are called for polls, but also they don't use social media, making it very difficult to predict what these people are going to do."
Predicting elections in other countries is a somewhat easier process since the national vote often determines the outcome. In the US, however, machines must be trained to learn different models for the electoral college that coincide simultaneously with the national vote predictions.
Mr Makse said AI models like his are still in the process of learning how to rescale and predict outcomes with sampling biases and other such limits, and that, one day, machines would be able to easily make up for that lack of knowledge.
Whether AI will correctly predict next week's presidential election remains unclear. But one day, AI will eventually be able to predict the outcome of the electoral college in any given scenario, according to Mr Makse: it's not a matter of if, but when.
"The main difference with traditional polling and artificial intelligence is that artificial intelligence is always getting better and better," he said. "Every election is a new data point, even if you didn't get it correct … you put all this information back in the model, and then it learns from every election."
Tags
Who is online
461 visitors
This I can see. The rural vote is very likely favorable to Trump. This indicates how the AI-based methods suffer from the same issue as conventional polling ... getting a representative sample. AI-based methods do not ask people questions so they are not limited by those who will volunteer to answer. But they are limited by those who engage online and even in the USA that represents a skewed sample.
This is going to be interesting. At least we can get that much out of this election.
The rural vote results in Arizona is going to be interesting. Mark Kelly holds a sizable amount of potential liberal Democratic votes in the large metropolitan areas of Phoenix, Tucson, Flagataff and other areas while McSally seems to hold more of the rural areas, which as stated tend to be conservative and/or Republican in makeup. Wonder what said AI makes of that?
Are there more rural or more metropolitan people?
Population wise, it may be a even split which is what makes it interesting.
Yes, technically Trump still has a chance.
yes, and I recall him having NO PATH TO 270 last election, too!
Trump was leading in many daily polls in swing states on election day - but the aggregate models weighted the lower reliability last day polls lower against the longer earlier trend where he had been behind. The same polls don't show the same movement this cycle - outside of a couple polls in Florida - which should be a toss up state - though even there the presence of significant mail in balloting will make late shifts interesting and even harder to predict than in 2016.
yes, but many, many talking heads proclaimed loudly and often that Trump had no path to 270--in other words, he didn't have a chance to win the election!
538 - for example - gave Trump a 1 in 3 chance of winning in 2016 vs. a 1 in 10 chance of winning in 2020. Definitely has a path - just even more narrow than it was four years ago.
He lost the election but won the electoral college vote. 2 different things.
I recall seeing many articles declaring that Trump simply had no path to 270, period.
https://www.politico.com/story/2016/06/donald-trump-path-to-victory-224239
They may be different to YOU.
if Hillary won the election, she would be sitting in the WH. That is how it works. She lost.
I know many people like to think the popular vote really, really means something, but many of them just don't understand how Presidents are elected.
They are different to anybody, except Trumpists who refuse to admit he wasn't elected by the people.
The election is what is going on now. People going out to vote.
AFTER the election, is when the votes are added and the Electoral College steps in. And as you have seen, the electoral college does not always represent the will of the people as shown in the election results.
I don't know how to explain this any simpler.
You may attempt to "explain" whatever you wish. Trump won the Oval Office just like EVERY President--by obtaining the required number of electoral votes necessary, as defined in the U.S. Constitution.
Now, if you can just point out where in the Constitution it describes the winner of the Presidential election being whoever receives the most popular votes, I will GLADLY look at it.
And since we both know you just can not do that, I would say the popular vote doesn't decide the election. Prove me wrong!
But if you want to hang onto your fantasy, I am cool with that. Me, I will stick to reality and the U.S. Constitution. But, hey, that's just me!
So, California's electoral votes did not reflect the will of the people of that state?Michigan? Texas? New York? Florida? Wisconsin? Pennsylvania?
Please present ANY evidence that any states' electoral votes did not reflect the will of the people of that state.
This should be fascinating!
Trump single handedley changed the entire election process in the US and stole the election.
And he miraculously did it BEFORE he was the President!
Some people just don't understand how Presidents are elected, or that the party who receives the most overall popular votes for Congresspeople and Senators doesn't necessarily mean they won the elections.
Which is exactly what I said, so why are you arguing about it?
My guess is that it galls you that he lost the election by around 3 million votes.
Why would I do something that is off topic from the discussion?????
Did you not read my comment? Why are you trying to argue something I never stated? What is wrong with you to make you so unwilling to stay on topic?
1st, no fantasy. Trump lost the 2016 election by 3 million votes.
2nd, you are not cool with it, or else you wouldn't try (over and over) to argue about something I never claimed. Are you that insecure about election results?
Deeeflection!!!
I see what you are doing, you cannot support your argument so you are trying to skew it to a different direction.
My claim is very very simple.
Now let's see if you can stay on subject, or if you're going to try and deflect again.
Then you made a terrible guess. I don't care about who got more popular votes, as that simply isn't the way Presidents are elected, so WHY would any person actually care about something totally meaningless? I am CERTAINLY not going to waste my time griping about who "won" the popular vote--I care about who wins the ELECTION, which was Trump in 2016.
Only one person here brought up the popular vote, so do tell me all about how I am the one insecure about election results--which clearly shows Trump won?
Think what you wish, but YOU are the one claiming that the will of the people was somehow thwarted, so if you truly feel that way, you should probably be able to tell me if the people residing in any state in the Union had their votes thwarted, based on which candidate received the most popular votes. I notice you refuse to answer that, so who really is deflecting here?
2016---Hillary Clinton had more overall popular votes, which is something I have never denied. It just doesn't fucking matter.
2016--the EC "awarded'" (read, voted in accordance with the majority of the residents of their respective states) the U.S. Presidency to Trump in accordance with our laws.
Now, let me ask YOU a couple of questions:
1. Who is the President? And did the President win the election as described in our Constitution or not?
2. Has any President in your entire life not been "awarded" the Presidency by the EC in exactly the same manner as Trump?
No deflecting now!
And you have yet to dispute that claim.
And once again you are trying to skew it into something else. Or are you claiming that a voter in one state should should have more control over who is elected than a voter in another state?
The winner will be determined by key votes in key states. The Electoral College was designed to protect and maintain the power of the status quo.
My example is this: Had the plurality of the vote count in 2000 determined the election there never would have been a 911, Iraq War, Homeland Security, Patriot Act, TSA or $150 OIL.
Really, so the Clinton CIA and intelligence community wouldn't have lost the organizing terrorists that entered and exited the US several times before Bush ever took office? I am sure Gore would have been quick to replace them all with his incoming administration./S
As for the rest Democrats judging by how many voted for the Patriot Act and Homeland Security- don't hold your breath.
I love revisionist history. Especially when it is so easily disproven.
As for your argument that the Presidency should be decided by popular vote- try making fly over country's votes worthless. Nothing will end the US faster than removing the voting voice of millions of US citizens.
another factor that will hurt the AI model is social media platform censorship, in order for it to predict accurately it not only has to have the information , it has to have all the information even that which is censored by platforms.
the only social medai i have is the book of face and even then its used only to stay in touch with friends and family , frankly my page is rather boring and dull......unless im being sarcastic and a major Ahole to someone on my friends list ( in a good natured ribbing way of course)
Mark,
The AI doesn't pay attention to social media.
Btw... I learned a new term today! It's called Election Stress Disorder. Do you think that might have to do with social media and a 24 hour news cycle?
maybe i misunderstood the article or am confusing it with another one posted on the site that the AI does comb social media .
as for the ESD, i think it is in full evidence here on the site . and yes i think social media and the 24 /7 news cycle can be blamed , we use to call it election burn out remember?
the point where one just wishes the whole damn thing was over .
Did you see this seed?
It is quite possible that it does not allow for the electoral college unpredictablity, and the other variable factors referred to in the comments above. However, it clearly pays attention to social media.
I get so tired of the endless campaigning! And the endless asking me for money!!!!!
It's also a platform that aggressively censors pro-Trump content. That seems like a terrible foundation for making predictions about what the American people want politically.
Tacos, the team actually said, they don't know. That they can see the shortcomings of the AI. How is that biased?
They’re trying to make some kind of scientific prediction based on a source known to be intentionally biased. That doesn’t make a lot of sense.
perri i think the point is , for the AI to accurately be predictive ( make a good prediction) all information needs to be input and let the algorhythm do what it was taught , to put in censored or faulty input will effect the algorhythm and thus the prediction in a skewwed manner .
grandpa was alive when computers took upo whole rooms and were very expensive , he said about them , a waste of money , if the information put in was wrong , because it would give you a wrong answer for what you needed , though the way he said it was " you put shit in for information , you will get shit out for an answer thats wrong