As a high school student in Silicon Valley, Ash Bhat’s entrepreneurial career began when he built and sold his first company, which focused on app development geared toward high school students.
To describe his projects from that time, which were similar to applications like Blackboard or Canvas used by many college students, Bhat says, “We were bringing together resources for high schools, so, for example, if you were at a high school, you would be able to turn in your homework, see bell schedules [and] read announcements. All of that from iPhone and Android apps. I was building a company around these apps and ended up getting acquired by a start-up.”
With respect to whether majoring in computer science was a foregone conclusion for him, especially given his prior experience, he shares that he didn’t actually choose to major in computer science (CS) but in interdisciplinary studies.
Ultimately, Bhat created his own major by combining philosophy, sociology and a bit of CS. He says, “One of the things I wanted to do, coming into Berkeley, was diversify different fields I have knowledge of and use to see the world. So, I took a few really cool CS classes, but I also took interesting social theory and philosophy classes as well.”
During the spring semester of his sophomore year in 2017, Bhat developed Presidential Actions — an app that pulls primary source data, such as executive orders, presidential memorandums and proclamations from whitehouse.gov, for both iOS and Android.
The impetus for Presidential Actions — the Milo Yiannopoulos protest in Feb. 2017, after which Trump had threatened to pull Berkeley’s funding — was Bhat’s introduction to politics.
He and his friends decided that if Trump was going to pull funding, they wanted to know as soon as it happened. Bhat’s response was to build a really simple application. In fact, it took him only 45 minutes to create and load it into the app store.
Literally overnight, the Washington Post picked up the story and published an article about it, after which Bhat had an OMG moment as he came to a certain realization.
He recalls thinking, “Politics have a lot of engaged people who are looking for solutions to the many problems that exist.” Today, news and politics represent areas in which Bhat finds technological solutions lacking.
Over the course of the next few months, Bhat and his lifelong friend, fellow Berkeley classmate and roommate Rohan, worked on random MAI projects together.
Then, in May, they saw an announcement on Facebook in which the social media company acknowledged the existence of fake news and pledged to solve the problem on their platform.
Ironically, right below Facebook’s news release was a fake news article about Obama. When thinking back to that day, Bhat says, “We were like, ‘Oh shit. They’re not doing enough about this.’”
That “oh shit” moment, which caused them to ask themselves how solvable this problem really was, turned out to be the genesis of their next big idea: NewsBotAI — a project created around the idea of detecting fake news on Facebook.
When discussing the insidious nature of fake news, Bhat notes that the issue with fake news is that it quite often makes sense or appears to be credible, and people tend to believe it is true without fact-checking it, especially if what they’ve read falls in line with their personal beliefs.
To describe how to check content, Bhat says, “Ideally, we should be checking all content, but that, right now, is very impractical.” No one could possibly fact-check every article they come across in a given day, as Bhat discussed in a recent TEDx.
So, Bhat and Rohan started looking into the concept, thinking it would be a cool project to develop — something they could get their friends to try out and give them feedback about.
Bhat says, “And we built out this app — a messenger bot that you could message. We put it on Facebook, primarily for our friends. However, when Mic decides to write about it, our user base blows up. We ended up with tens of thousands of users who started using it every day to classify their news stories. It was another ‘oh my god’ moment. It was just this repeated pattern of us putting out a solution to a problem and getting this amazing response. There was clearly something here.”
To offer a bit more detail about how NewsBotAI works and detects fake news, Bhat describes it as a statistical modeling of news stories that examines various attributes of the article and its associated news site — author, word count, types of words used and the number of articles on the site — and turns it into a set of statistics.
Reviewing fake articles, real news and pieces that exhibited partisan bias allowed them to characterize the patterns presented in each of these types of pieces.
From there, Bhat says, “You essentially have this really complex statistical model that you toss an article into and turn the article into just a bunch of numbers, and it spits out a classification, so it will say, ‘this is fake news,’ ‘this is a right-leaning article,’ ‘this is a left-leaning article,’ ‘this is satire.’”
So, not only does it identify fake news, it also identifies ideological slant in news, which Bhat claims the majority of people use it for.
As Bhat and Rohan launched NewsBotAI, they also started speaking with different media companies — one of which wanted them to join their team — and realized that there was a lot of interest in these types of projects.
As a result, the two old friends co-founded the company they cleverly named Robhat Labs (pronounced Robot). Also at the same time, “Wired” began writing a story about their journey and the process surrounding building applications that address fake news and the spread of misinformation.
Realizing how much fake news was originating on Twitter via automated accounts known as bots, Rohan and Bhat turned their attention toward the development of a project capable of detecting fake bots on Twitter.
BotCheck.me was the result. While they would have liked to have built something directly with Twitter, Bhat notes that it is incredibly difficult to actually get a deal with the social media platform.
With that in mind, they opted instead to build a Chrome extension that Mac or PC users could download and install to add an extra button to their Twitter feed.
In addition to the standard Twitter buttons — such as the retweet, heart and report button — users will also have a BotCheck.me button, which lets them know if the tweet in question is from a bot or human user. There are plans to expand the use of the app to Reddit, which Bhat explains is another hotbed for bots and fake news.
BotCheck.me and the “Wired” article, both ready for launch at the same time, took things up another notch. With two important things happening at once, Bhat says, “and then that blows up to another level that we hadn’t experienced before.”
The response they saw was a mixed bag. Bhat says, “All of a sudden, we were seeing conspiracy theory videos of us appearing on YouTube, and we have hundreds of people tweeting at us every single day. Hundreds of thousands of people have seen or used our product…We didn’t expect the response we saw.”
Still, there were accolades to be had, which led to new opportunities to explore. E-zine “AdAge” compiled their “Creativity 50,” a list of the 50 most creative people of 2017.
As he remembers the surreal moment he and Rohan saw the list, Bhat says, “It was Stephen Colbert, Rhianna and then us. And we’re like, ‘What the heck?’ At this point, we were both surprised and incredibly humbled.”
After “Creativity 50” was released, they were able to leverage the high level of recognition they were receiving to begin to “deal with different political groups.”
Bhat noted that he couldn’t really comment much on their current work, but he says, “[We] want to make sure the 2018 elections aren’t as affected by fake news as the 2016 elections were, so we are working with a few political groups now to make sure that doesn’t happen.”
Bhat and Rohan are driven by a sense of purpose: transparency surrounding news on social media platforms. According to Bhat, Robhat Labs perceives fake news as “analogous to a virus or infectious disease in the sense that you have to start with detection, then move onto treatment and then special intervention.”
Currently, their technological solutions function in the capacity of fake news detectors that allow users to proactively validate news and news sources.
Ultimately, however, they don’t view this as a “scalable solution,” as the onus for utilizing these tools to identify fake articles or social media bots is entirely on users themselves. To address this, Bhat says, “So, we are looking at other solutions as well [and] working with bigger organizations so that they can filter out fake news on behalf of users before users see it.”
As for the future of Robhat labs, the vision is as clear as their current sense of purpose. Bhat would like to follow their current model from detection to treatment and prevention, with the goal of preventing the proliferation of fake news and misinformation within the next five years.
He expects this problem to continue to grow and believes what people are seeing now is just the tip of the iceberg, as far as the effect it is going to have.
School, work and life in general
A valuable benefit of studying at UC Berkeley has been having access to truly amazing mentors. Bhat describes the machine learning and MAI scene as “super vibrant,” asserting that the field has some of the best professors.
He says, “I had some super influential professors, from my social theory professor to our data science professor, Joey Gonzalez. He’s just a super accomplished guy, and at the same time, he’s so humble. I think he sold a company for $200 million to Apple once. Just learning from him was so incredible in terms of experience. He knew ‘This is going to happen; watch out for this.’ The mentors I’ve had in my life are definitely a big reason why I’m where I am today.”
With respect to how he and Rohan juggle life, school and career, Bhat mentions that they keep themselves as organized as possible by proactively managing their calendars so they don’t have to attempt to remember every meeting or task they have.
Additionally, documenting everything that comes to mind is a huge part of the organizational process of the Robhat Labs co-founders.
Thoughts and ideas are processed as they come to mind, and both make it a point to sit down and write essays or two-pagers to ensure they don’t lose any information before the opportunity to work on new ideas arises.
Despite the highly organized way he manages both the business and his education, the explosive speed at which Robhat Labs took off over the past year, coupled with school, became more than Bhat could juggle at one time. As a result, he’s taking a break from classes this term.
As far as his future for education goes, he feels this has been his most research-intensive semester to date. There is the opportunity at this point to pursue a graduate degree, and he is keeping his options open.
He declares, “At the end of the day, the big thing is to continue learning.” He feels that Robhat Labs, as it creates technological solutions to oppose fake news, is at the forefront of this field right now and is a great learning experience in and of itself.
Work-life balance is something Bhat says was lacking when he started his first company in high school and, having learned from that experience, it is immensely important to him this time around.
As he helps to build this new company, he has been careful to begin applying principles he deems essential to both success and well-being, and practicing work-life balance is one of the principles for him.
Time to decompress and reboot is vitally important, and he accomplishes that by spending time with his friends and ensuring he and his girlfriend take time to go on dates.
As our conversation drew to an end, I asked Bhat if there was anything else he would like to share with other college students. He says, “I think one of the biggest things I do is I write pretty voraciously whenever I need to think through a concept. I think we are pretty limited in terms of what we can think without writing because writing just gives you so much more mental capacity. So, whenever I have to work on a long or hard decision, or I’m stressed out about something, I write it out, and it sort of helps me think.”
Growing up in Silicon Valley, Bhat feels he and Rohan have had the benefit of access to “unfiltered information.” He sees this new age of disinformation as a damper on the huge opportunity provided by the internet to learn anything.
Building protections around and being able to secure public access to information online has been a huge motivation for him.
So, while he isn’t sure where this path will lead him in the future, he knows that helping users assess the veracity of information and promoting access to information will be a core component of whatever he does.
While it wasn’t Bhat’s goal when he first started out at UC Berkeley to design technological solutions to political or news related problems and create the now famous company Robhat Labs, it has since become his driving force.
After all, he says, “For me, it’s not just an interesting business; it’s also an ethical and moral issue that I want to be making a difference in. In terms of that, it’s a good problem for me to put a lot of time into and help solve.”
