The Future of Verytas

Posted by Sam Mallikarjunan on Mar 24, 2016 9:55:21 AM

Below, you can read the expanded Verytas pitch from Indiegogo and learn more about our vision for the future of Verytas.

You: Ok, you got me to click. So what problem are you trying to solve and why should I care?

Verytas: Creating and sharing poorly researched -- or even entirely fake -- content on the internet is becoming a better business model than well researched journalism. It's cheaper to produce, and spreads faster than ever through social media. Traditional journalism requires research, citations, and a series of checks and balances that makes their writing slower and much more costly to produce. We want to make it less profitable to spread fake news and cheaper to produce well-researched content to start bridging the gap in the two business models.

You: You want to tackle misinformation on the internet? Are you insane? 

Verytas: We're insanely optimistic! First, there’s a misconception that you have to have a complete and unassailably perfect solution before launching something. In reality, no one starts off by building The Perfect Thing. You build a Thing that solves for a specific use case so that you can learn and chip away at the larger problem. There are challenges with misinformation that are actually really easy to solve (identifying satire sites, flagging things that are objectively true or false) and then there’s a limitless universe of gray-area problems that we’ll chip away at over time. Just because we can’t solve the entire problem right now doesn’t mean we shouldn’t try to knock out the easy use cases we can solve.

You: Ok, but I already research everything I see online to check if it is true or not. Why should I care?

Verytas: Nonsense. No one has that kind of time. Even professional journalists, officials, and other people who should know better get fooled by content online. You can take our quiz to see how good you are at recognizing fake news. If you don’t get 100% right, then you have nothing to lose by simply supporting a tool that gives you access to more information easier. If you do get 100% right, then you have a moral duty to help others have access to your great wisdom, so contribute anyways :)

You: Ok, how does it work?

Verytas: Verytas is currently a Chrome-only browser extension that reads the links, text, and images on your screen and checks to if any community members have added links to the original source material that the author was referencing, or links to citations disproving the content.

For example, the Verytas Chrome extension is currently able to see the links that appear in your Facebook or Twitter feed and check them against our list of hundreds of satire and fake news domains (no data is transmitted about what you're seeing or clicking, it's all done privately inside the extension). If a link is a match, Verytas highlights that link in purple:

 

sample satire article highlighted purple in a Facebook newsfeed

The next step for Verytas is to go beyond obvious satire and fake news and help you do your own research faster. Community members will be able to add links to citations relevant to the article or image and those will appear right in your Facebook or Twitter newsfeed. This lets you, for example, read the study that an article is supposedly citing to make sure that it says what the article says it does -- without having to track down the original source yourself.

Instead of purple, a link will be highlighted in green if community members have added citations that support a claim in an article by highlighting the text on the article itself and adding their citation. The link in your newsfeed may also be highlighted in red if they've added citations that disprove it, or in yellow if community members have added citations that both prove and disprove individual assertions made in the article.

You: How do you know if something is true or false? What if someone posts misleading citations because people are jerks and everybody lies? Who fact-checks the fact-checkers?

Verytas: Despite the name, we’re not trying to tackle the philosophical definitions of “truth”. We’re taking on a huge and complex problem by starting extremely simple: The only thing we’ll show you is if there are original source citations available that either prove or contradict specific assertions (such as data from a study or video of a statement made by a politician). Verytas is here to help you scale your skepticism, not do your thinking for you.

That disclaimer aside, citations added to Verytas will be scored based on our algorithm which primarily factors the strength of the source being used and the history of the Verytaser. If you include a link to a domain people have approved for citations before and you’re brand new to Verytas, your citation will be shown. But if Walter Cronkite himself comes back from the grave and tries to cite The Onion, his submission will not show up. Information cited will be placed in a queue to be checked by others, and the more Verytasers that confirm that citation the stronger the citation will be weighted to determine if it will be displayed to users or not.

The algorithm and code will be open-source and available for people to review so that you know we’re not up to any funny business due to our own biases. Also, for any links or images for which there are strong citations both contradicting and supporting an assertion, we’ll default to showing both and let you decide which to believe.

You: Okay, but you mentioned memes. A lot of what I see on Facebook and Twitter is people sharing quotes or charts or other memes with data but I have no way of knowing if they're true or not.

Verytas: That's a harder problem to solve, but it's not impossible. Articles are relatively easy, since it has a single URL we know that if Walter adds a citation from this URL that we should show it to you if that same URL appears in your newsfeed. But image matching technology has actually gotten pretty good. We're currently planning to use TinEye's API to help us check if an image has a match in our database. It won't be as perfect as the URL matching, but it should save you a lot of time. Again, when in doubt, we'll default to showing you any citations we think match and let you decide.

You: You can't possibly fact-check the entire internet. How will Verytasers know what people want them to check?

Verytas: You're right, of course. Although Verytasers can add citations to anything they want, there will be a queue that users can submit to to have items they're interested in checked. Normal users can submit an image, a snippet of text, or an entire article and they'll receive a notification when citations are added.

You: Who's going to do all this work? No one is going to spend time proving things wrong on the internet.

Verytas: Have you never been on Reddit? There are already people doing detailed research of varying quality to win arguments online. But even when someone finds something solid relevant to an article or assertion, that research is lost (let’s be honest, they’re not convincing the person they’re arguing with). In addition, sites like Wikipedia have shown that there are tens of thousands of people in the world who care about information and knowledge enough to volunteer their time. No one has time to check everything they see online -- not even you RedditFactChecker127. But many people working together can reduce the turnaround time from requests incoming to research outgoing enough to be useful. Also, as we'll explain later, this is hopefully something you'll be able to make good money doing.

You: Okay, I get it. People can add citations to articles or images and if they're strong sources or if they're verified by other users they'll show up if I view the article or if the article appears in one of my social feeds.

Verytas: Yep. You got it.

You: So if a writer or a journalist doesn't have time to add citations for every assertion they make in an article, you're going to make it even harder for them to get readers. You've taken an industry that was already competitive and hard to make a profit in and made it even harder. Are you proud? Do you feel good about yourselves?

Verytas: You are not wrong. That's exactly what we've done -- we've made it so that articles that use weak or no citations in their content will hopefully get less traffic. But we're not stopping there, we also want to make it easier to create better content.

You: ORLY? Tell me more.

Verytas: We've interviewed dozens of journalists, writers, and reporters -- and you're right. A huge problem is that they simply don't have the time to dig and do research before they have to publish. They have to keep up with an incredibly fast and ever-accelerating pace of digital publishing deadlines. However, we've also found that there is some consistency in the types of data they need and where they find it. So we're also building research enablement tools that make it easier to find citations for the content you're writing.

You'd be shocked by how many writers still have to go to a website and use whatever terrible site search functionality that site has in order to dig through their data to find research. Many of these sites have APIs or structured data already, so Verytas will collect those sources of research and show them right on your screen when you're writing. 

For example, if you're writing an article about a specific congressperson's vote on an oil pipeline, Verytas will go to OpenSecrets.org for you and check to see if they have any campaign donors from the energy industry. If they do, Verytas will pop up a notification telling you that we've found research you should consider.

This is something we'll be continually adding new sources for, so the more people submit citations manually the better we'll be able to know what we should automate and make easier.

You: Okay, that's kind of cool. It has potential anyways. But it's still the writer having to chase down a bunch of research themselves. Surely there's a better way?

Verytas: There is! It's as if you and I are the same person having some weird conversation with ourselves.

We don’t just want to make it harder for misinformation to be a profitable business, we also want to make it more cost-effective for media companies that want to share accurate information relevant to the public interest (yes, they exist).

We know that research is hard and time consuming. We also know that just because you're a great writer doesn't mean you're a great researcher -- and vice versa. The great thing about the internet is that there's almost always someone who's awake and available right now who's better than you are at some specific task, so Verytas is also building a market network to combine the two.

This video explains more:

If you'd like to see a first draft of the product demo video we've made for the market network, you can click here.

This will also help make participating in the Verytas community more valuable to our members. As requests for research come in, they'll go to the Verytasers with the best reputation for fact-checking based on our algorithm first. So participating in fact-checking content from social media boosts your score (if the citations you submit are verified by other community members) and you'll get first crack at any paid jobs that come through!

There are lots of people to whom fast, accurate research is valuable -- not just journalists. And our sales and marketing teams will be hard at work matching them with you (once we hire them).

You: I get it. So Verytas doesn't just make it harder to get traffic to fake articles or articles with no citations, it actually helps writers and journalists to create better content faster or more cost effectively. Like an Uber for internet research.

Verytas: Ugh, we're so glad you said that and not us. We all roll our eyes when we hear the "Uber for X" startup cliche, even though it does a good job of conveying how a startup concept can provide a higher quality experience at a lower cost to end users by making it so that the top providers have no down time and therefore make more per hour even though they make less per task.

You: What if I think I've done a good job writing an article and don't need more research, but I want help checking it without that information being shown to the public? I'm not perfect, but I also don't want to get a reputation for poorly cited content.

Verytas: No worries! Another feature of the Verytas market network will be the Content Quality Assurance subscription. For a monthly fee, you can post your content there before you publish it and Verytasers will be able to send you "bugs" they think they've found. These can be factual errors, citation errors, or even something as simple as grammatical or spelling errors. If you accept the errors that people find, that Verytaser will get paid a fee, but you don't pay anything additional. This way, the Verytaser knows that if you're rejecting a citation it's for good reason -- not just to save money.

We're sure we've made some grammarical errors in this conversation already, and it sure would have been great to have had hundreds of people tear it apart quickly and cost-effectively before the grammar police got their hands on it.

You: What about citations someone has already added to an article I've written? Can I pay to have those removed?

Verytas: Right now, no. Once a citation has been added to a public article, no amount of money can cause it to be removed. If you publish an update to the article, the citation will no longer appear. Any Verytasers who had previously added citations will get a notification so that they can see what you've changed.

You: If I contribute to this campaign, how will my money be used?

Verytas: Any funds we raise will be used for developer time and relevant technology that we need. For example, we’re using an image matching service called TinEye to match memes you want to Verytas with memes in our database. It only costs about $0.01/search, but those pennies add up! We also need to pay some more developers to help us get off the ground and there are additional technology expenses. The money you give us will go right to that (we’re not getting paid from this -- we have day jobs).

You: It seems like there's a real business model here. If Verytasers are going to be able to make money there's obviously potential for you to make money as a startup. Why are you crowdfunding instead of raising more traditional venture capital?

Verytas: We did have meetings with several really smart VC's who gave us a ton of excellent and useful feedback. That being said, most VC's aren't here to change the world, they're here to get a return on their investment portfolio. While there's definitely a business model here (we strongly believe that lasting and scalable change requires a revenue model and can't just rely on donations), some of the features that are really important to us won't generate revenue or capture a share of a specific "market" (like the social media newsfeed analytics). Smart investors would want to steer us away from those "fascinating distractions", but we think they're actually core to our long term growth and value. While we prove that, we plan to run under our own steam so that we're not beholden to any investors.

You: Who’s behind Verytas? Your website is terrible and I can’t tell who you are. You could be aliens plotting to destroy Earth!

Verytas: Our website is terrible. We've been running this as a side project for a while, and we haven't invested much time into making it beautiful. As for who we are, right now the Verytas team has two members:

Sam Mallikarjunan is the co-founder who handles the marketing and other business related functions (such as crowdfunding campaigns). His day job is at a software company in Boston.

Andrei Oprisan is the co-founder who handles the engineering and technical requirements of Verytas. He’s the one actually building this. His day job is at that same software company in Boston.

You: Logic, reason, moral obligation, and a hope for a better tomorrow haven’t motivated me to contribute yet. What else ya got?

Verytas: Here are our co-founder’s two dogs (Simon and Jayne, named after characters from Firefly) judging you harshly if you don’t contribute so that we stop spending their treats and toys money on Verytas:

You: Others have tried to build technology to auto-fact-check the internet and so far have failed. What makes you think you'll be different?

Verytas: Our approach is fundamentally different than, say, what the folks at Google or Facebook have tried to do so far. They're trying to develop programmatic solutions to this problem everywhere for everyone on the internet in real-time. Verytas looks at this more like a manufacturing problem: We take inputs on what people want checked and try to minimize the turnaround time to get people the citations they need.

And, again, despite what the name Verytas implies, we're not trying to solve for a Grand Unified Theory of Truth. We've actually built it into our strategy to perform GUTT checks and make sure we're not even trying to. So while, yes, the concept of flagging everything as either "true" or "false" is a lot more complicated than people think, simply giving people easy access to the most credible original source citations is a pretty good start.

The "right" answer for this problem will problem be somewhere in between a programmatic and community driven solution -- and it will be a never ending process as spammers and scammers continue to try and game the systems we build.

You: Fair enough. Will Verytas do anything else?

Verytas: Yes! There's one more feature we think you'll love if you've read this far: Verytas News. You're going to hate getting this email in your inbox every morning. Verytas News sends you the most well-researched and best cited content on the web that's the opposite of what you currently believe on a given topic. 

Basically, this is the exact inverse of how Google Now "popular with readers of..." cards work. Whereas Google Now shows you other articles that are popular with other readers of the site you read, Verytas News will show you the content with the most supporting citations that backs the opposite of your current position. If you're pro gun control, for example, and an anti gun control article comes out that gets lots of supporting citations from the community, you'll get that in your inbox.

If anything, you may want to contribute to our campaign just to make this newsletter a reality.

You: Won't that just make me angry and want to go add refutation citations to that article?

Verytas: Yes! In addition to taking a sledgehammer to your news and information echo chamber, Verytas News will hopefully encourage more citation contributions from community members and help create rich, balanced articles with a variety of citations. Rage-powered product adoption, for you startup nerds out there.

You: Alright, you've convinced me. I still think this problem is too large to actually solve, but I'm willing to chip in $10 or so to give you all a chance to make a dent.
 
Click Here to Support Verytas

 

Topics: Startups, Technology, News

Verytas is a tool that lets everyday people easily fact-check the content they see on social and news media sites like a pro.

Verytas pulls from commonly used professional sources as well as human-generated citations or refutations and automagically overlays the information in real-time on your normal browsing experience.

Subscribe to Verytas

Recent Posts