<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Marina Walker Guevara on Medium]]></title>
        <description><![CDATA[Stories by Marina Walker Guevara on Medium]]></description>
        <link>https://medium.com/@marina.walker?source=rss-184b8833d2fe------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*C2MxpO460GrlLjckQEhEKw.jpeg</url>
            <title>Stories by Marina Walker Guevara on Medium</title>
            <link>https://medium.com/@marina.walker?source=rss-184b8833d2fe------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 02:03:53 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@marina.walker/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Five ideas we need to ‘unlearn’ to build resilient media organizations]]></title>
            <link>https://medium.com/jsk-class-of-2019/five-ideas-we-need-to-unlearn-to-build-resilient-media-organizations-7316af469877?source=rss-184b8833d2fe------2</link>
            <guid isPermaLink="false">https://medium.com/p/7316af469877</guid>
            <category><![CDATA[investigative-reporting]]></category>
            <category><![CDATA[journalism]]></category>
            <category><![CDATA[nonprofit-media]]></category>
            <category><![CDATA[organizational-culture]]></category>
            <category><![CDATA[collaboration]]></category>
            <dc:creator><![CDATA[Marina Walker Guevara]]></dc:creator>
            <pubDate>Mon, 01 Jul 2019 23:18:58 GMT</pubDate>
            <atom:updated>2019-07-02T17:53:33.547Z</atom:updated>
            <content:encoded><![CDATA[<p>What a dance class at Stanford taught me about destabilization, chaos and journalism.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SBzTjQ4yxuTwWFIIHCCd4w.jpeg" /><figcaption>Choreographer Alex Ketley teaches the popular Stanford dance class “Hot Mess.” Photo: Andrea Basile</figcaption></figure><p>“A dance class in how we become the worst dancer possible.” The description of this Stanford offering caught my eye. “As a class, we purposely explore chaos, failure, and ‘bad’ dancing, with the hope that then we will have a greater chance to understand and refine our personal notions around beauty.”</p><p>I spent a good chunk of my childhood trying to become a good ballet dancer. To fit that ideal of the ethereal ballerina, to master those wicked turns and leaps. As an adult, Argentine tango, another highly prescriptive dance with crystal-clear notions of good and bad, became a new goal and obsession.</p><p>And here I was, some years later, enrolled in a class called Hot Mess where the goal was to ‘unlearn’ those ideas of beauty and good dancing and explore what can happen when we allow ourselves to be lost.</p><p>As we ran across the studio improvising moves or sounds or infusing our ‘dancing’ with phrases that made no sense (birds! ice-cream! shrimp!), it became clear to me that there were two well-defined groups of students in the class: those who were having the time of their lives; and those like me who were excruciatingly trying to unlearn decades of “proper” dancing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/604/1*uXagK5qPWbCSTc-h81UrYw.jpeg" /><figcaption>Tango dancing in the streets of Washington DC.</figcaption></figure><p>On my bike ride home after class, I started to draw parallels between my experience at the dance studio and journalism. Our profession, too, is in the midst of a great deal of chaos, transformation and promise. There are forces at work — economic, technological, political — whose course we can hardly predict or control. And then there’s us, journalists, reinventing our profession and the business models that might sustain it into the future, experimenting, distinguishing shining objects from those with true value, acknowledging that we need others — different perspectives, experiences and skills — to build something that lasts.</p><p>There is so much to be hopeful for.</p><p>Earlier this year, I helped judge the <a href="https://aiethicsinitiative.org/challenge">AI and the News Open Challenge</a>, which funded creative ideas at the intersection of artificial intelligence and the news. I was impressed with the inventiveness of many of the more than 500 projects submitted. In the U.S. and worldwide, <a href="https://inn.org/wp-content/uploads/2018/10/INN.Index2018FinalFullReport.pdf">the nonprofit news space</a> continues to grow, with <a href="https://gijn.org/member/">charity and community supported newsrooms</a> popping up at a fast pace in response to shrinking media freedoms and expanding news deserts. <a href="https://www.icij.org/blog/2019/04/how-panama-papers-the-story-too-big-for-one-news-outlet-changed-these-reporters-lives/">Collaboration</a> powered by data and technology has forever changed the way investigative journalists work, creating opportunities for impact at a scale never seen before.</p><p>It’s a time of destabilization, learning, fear and opportunity in journalism.</p><p>Yet, how are we journalists tackling it to make sure our organizations last well into the future? What assumptions and mindsets are we bringing to these new journalism spaces?</p><p>Here’s my take on five ideas we need to <em>unlearn</em> as an industry if we want to seize the potential of our own <em>hot mess </em>moment:</p><ol><li><strong>Our job is just to tell the story.</strong></li></ol><p>It’s more complicated than that. We need to be intentional about the impact of our stories. In other words, be clear about why we are doing what we are doing and what needs to change in the world as a result of our journalism. This mindset has big implications for story selection, for the way we report our stories (collaborating with other newsrooms every time it makes sense and with the communities we serve always) and for the resources we invest to stay with a story over time. At the International Consortium of Investigative Journalists, we are still <a href="https://www.icij.org/investigations/panama-papers/panama-papers-helps-recover-more-than-1-2-billion-around-the-world/">tracking the money</a> governments around the world are recovering as a result of the Panama Papers ($1.2 billion so far is the conservative estimate) and continue to report about the developments of investigations we did six years ago, such as <a href="https://www.icij.org/investigations/panama-papers/tax-raids-across-germany-sparked-by-icij-revelations/">Offshore Leaks</a>, which recently sparked multiple bank raids in Germany. Along the way, we have presented our findings in parliaments, met with high-school students to talk about the inequities caused by tax havens, and published an <a href="https://offshoreleaks.icij.org/">interactive database </a>of offshore companies that allows citizens to do their own exploration of the data that underpins our work. When one of our colleagues was murdered in Malta, many of our partners <a href="https://forbiddenstories.org/case/the-daphne-project/">joined forces</a> to continue her work.</p><p>2.<strong> Culture scales automatically.</strong></p><p>Silicon Valley, where I have spent a year as a John S. Knight Journalism Fellow, is full of horror stories of startups that imploded as they grew fast and lost their values and identity along the way. Curt Coffman and Kathie Sorensen, authors of “Culture Eats Strategy for Lunch,” argue that culture “is not the enemy of strategy and performance but an equal player in the game” and an essential competitive advantage. “Our competition can copy our products, methods, and service, they can hire our stars, duplicate our marketing, use our vendors, lure our customers with amazing offers, but they can’t clone the culture that created those advantages.” As our journalism startups grow and get more attention and funding, let’s remember the foundational days — the energy, the focus, the sense of mission — and let’s see that our core values are lived out every day in our newsrooms. No tech, cool innovation or branding can make up for a mediocre culture. Sooner or later, that bad culture will catch up with your strategy and crash it.</p><p>3.<strong> Journalism is an unhealthy profession.</strong></p><p>How we take care of our people is exhibit one of the culture of any organization. In my job leading global investigations, I come across an ever-larger number of journalists who tell me, quietly, how physically and mentally exhausted they or their colleagues are. It’s not just the daunting events they witness in their reporting, the physical and online harassment they face routinely or the debilitating lawsuits they battle. Now it is also the worry about the stability of their jobs; the pressure to raise funds to support their work and the work of the people they lead; the apparent inescapable reality of having to play multiple roles in journalism startups or in decimated local news outlets. The good news is that journalism-induced burnout has become less of the taboo than it was, and more journalists are <a href="https://www.cjr.org/special_report/burnout-journalism.php">talking candidly</a> about their experiences. “I am tired. I can’t continue doing investigations and raising money. It’s not humane,” said Mónica González, one of Latin America’s leading investigative editors, in a <a href="https://www.latercera.com/reportajes/noticia/examen-conciencia-monica-gonzalez/669890/?fbclid=IwAR1In4_W7JAWiuC2oepBkMeQBclicT7MgaiZ67fAyArakQHsC14tzJgSfK4">recent interview</a> as she stepped down from Ciper, the nonprofit investigative news outlet she founded 12 years ago. Unlike Monica, many other journalists would not know or admit that they, too, are at the end of their rope, as they model unhealthy behaviors to younger generations of reporters. <a href="https://news.ku.edu/2015/03/23/study-shows-journalism-burnout-affecting-women-more-men">Studies have shown</a> that female journalists are burning out at higher rates than men.</p><p>Journalism leaders, board members, funders, schools and professional associations need to take on newsroom burnout head on before it consumes our newsrooms and our business.</p><p>4.<strong> We must adopt/develop/fund the latest shining tech tool.</strong></p><p>“What kind of world do we want to live in and how can technology serve it, rather than drive it?” asked NYU professor Kate Crawford, one of the leading thinkers on the social implications of artificial intelligence, at a recent Stanford event. We can easily extrapolate that question to journalism. We grapple with it a lot in the field I know best, investigative reporting, where dozens of tools have been funded and developed, but only a few have scaled and made a difference, with <a href="https://www.documentcloud.org/">Document Cloud</a> being one of those successful examples. Not long ago, a group of journalists and technologists met to discuss the creation of an all-encompassing “super tool” for investigative journalists that would rival those used by the intelligence community. The project had a price tag of nearly $20 million, which has not been, and likely will not be, raised any time soon. To complicate things more, big tech companies and platforms are stepping in and offering journalists their own tools as well as training and funding.</p><p>Perhaps a more reasonable strategy, following Crawford’s logic, is for journalists to take a step back from all the confusion and the panic and ask ourselves what problems we are trying to solve with technology, what “pain points” in the reporting process we believe technology can help us address and what tools already exist out there that we can adapt for our own purposes without compromising our data or our independence. During my year at Stanford, I have studied how newsrooms can effectively partner with academia to use <a href="https://medium.com/jsk-class-of-2019/we-cracked-the-panama-papers-with-400-human-brains-can-ai-help-us-next-time-1f97164742a0">machine learning techniques in big data investigations</a>.</p><p>5.<strong> We can do this on our own!</strong></p><p>In this area, we have come a long way since the days when investigative reporting was largely a lone-wolf affair. Working together across newsrooms and borders journalists have reported stories that <a href="https://www.bbc.com/news/world-asia-40750671">toppled governments</a>, <a href="https://www.europeanpressprize.com/article/money-laundering-danske-bank/">exposed corruption on a grand scale</a>, prompted regulators <a href="https://www.nytimes.com/2019/04/05/world/europe/france-breast-implants-banned-cancer.html">to ban harmful products</a> and <a href="https://www.occrp.org/en/corruptistan/azerbaijan/khadijaismayilova/">continued the investigations of their jailed colleagues</a>. Residual reluctance to collaborate with others remains in pockets of the media, most notably in the United States. One can only wonder what the impact of certain stories such as the crisis on the U.S. southern border, Trump’s business deals around the world or the contamination of drinking water in the U.S. would have been had journalists collaborated more deeply with their peers at home and abroad. At the end of the day, it comes down to the impact we want our work to have on society (see point 1!).</p><p>Scientists just gave us a powerful example of what happens when we let go of our fear to share our work with others for the sake of something bigger than ourselves. It had been impossible until now to photograph black holes because they are so big and so far away. Scientists would have needed a telescope the size of the Earth. So instead, more than 200 scientists coordinated radio telescopes in different locations around the world to act like one giant telescope. As Seth Fletcher of The New York Times wrote, they had to adjust the telescopes to work together; they had to get the timing right; they had to hope that the weather would cooperate in multiple places. “Oh, and they had to hope nothing broke. It sounded crazy, but it worked.”</p><p>The premise for my Hot Mess dance class at Stanford also sounded crazy (“Dancers confront a number of movement and vocal prompts that are literally impossible to execute in any good way”). But it also, to my huge surprise, worked. I saw beautiful, expressive and powerful dancing as classes went by and we let down our guard. A new kind of beauty. Messier, less predictable, more inclusive and more authentic.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7316af469877" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jsk-class-of-2019/five-ideas-we-need-to-unlearn-to-build-resilient-media-organizations-7316af469877">Five ideas we need to ‘unlearn’ to build resilient media organizations</a> was originally published in <a href="https://medium.com/jsk-class-of-2019">JSK Class of 2019</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[We cracked the Panama Papers with 400 human brains. Can AI help us next time?]]></title>
            <link>https://medium.com/jsk-class-of-2019/we-cracked-the-panama-papers-with-400-human-brains-can-ai-help-us-next-time-1f97164742a0?source=rss-184b8833d2fe------2</link>
            <guid isPermaLink="false">https://medium.com/p/1f97164742a0</guid>
            <category><![CDATA[panama-papers]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[collaboration]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[investigative-journalism]]></category>
            <dc:creator><![CDATA[Marina Walker Guevara]]></dc:creator>
            <pubDate>Tue, 19 Mar 2019 20:22:41 GMT</pubDate>
            <atom:updated>2019-03-19T20:22:41.588Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a5N4vokpfnlmUnnJMou9zw.jpeg" /><figcaption>Credit: <a href="https://unsplash.com/photos/3y1zF4hIPCg?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Hans-Peter Gauster</a> on Unsplash.</figcaption></figure><p><strong>A new partnership between journalists and Stanford machine learning scientists aims to enhance the investigative reporting process. Here’s what we learned so far.</strong></p><p>As we approach the third anniversary of <a href="https://www.icij.org/investigations/panama-papers/">Panama Papers</a>, the gigantic financial leak that brought down two governments and drilled the biggest hole yet to tax haven secrecy, I often wonder what stories we missed.</p><p>Panama Papers provided an inspiring example of media collaboration across borders and using open-source technology at the service of reporting. As one of my colleagues put it: “You basically had a gargantuan and messy amount of data in your hands and you used technology to distribute your problem — to make it everybody’s problem.” He was referring to the 400 journalists, including himself, who for more than a year worked together in a virtual newsroom to unravel the mysteries hidden in the trove of documents from the Panamanian law firm Mossack Fonseca.</p><p>Those reporters used <a href="https://source.opennews.org/articles/people-and-tech-behind-panama-papers/">open-source data mining technology</a> and <a href="https://neo4j.com/blog/icij-neo4j-unravel-panama-papers/">graph databases</a> to wrestle 11.5 million documents in dozens of different formats to the ground. Still, the ones doing the great majority of the <em>thinking</em> in that equation were the journalists. Technology helped us organize, index, filter and make the data searchable. Everything else came down to what those 400 brains collectively knew and understood about the characters and the schemes, the straw men, the front companies and the banks that were involved in the secret offshore world.</p><p>If you think about it, it was still a highly manual and time-consuming process. Reporters had to type their searches one by one in a Google-like platform based on what they knew.</p><p>What about what they didn’t know?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5c8GeqvE8U0FqEy-mVCbjw.jpeg" /><figcaption>Panama Papers reporters work together at the German newspaper Süddeutsche Zeitung, which first obtained the data.</figcaption></figure><p>Fast-forward three years to the booming world of machine learning algorithms that are changing the way humans work, from farming to medicine to the business of war. Computers learn what we know and then help us find unforeseen patterns and anticipate events in ways that would be impossible for us to do on our own.</p><p>What would our research look like if we were to deploy machine learning algorithms on the Panama Papers? Can we teach computers to recognize money laundering? Can an algorithm differentiate a legitimate loan from a fake one designed to shuffle money among entities? Could we use facial recognition to more easily pinpoint which of the thousands of passport copies in the trove belong to elected politicians or known criminals?</p><p>The answer to all of that is yes. The bigger question is how might we democratize those AI technologies, today largely controlled by Google, Facebook, IBM and and handful of other large companies and governments, and fully integrate them into the investigative reporting process in newsrooms of all sizes?</p><p>One way is through partnerships with universities. I came to Stanford last fall on a <a href="https://jsk.stanford.edu/fellows/class-of-2019/marina-walker-guevara/">John S. Knight Journalism Fellowship</a> to study how artificial intelligence can enhance investigative reporting so we can uncover wrongdoing and corruption more efficiently.</p><h3><strong>Democratizing AI</strong></h3><p>My research led me to <a href="http://ai.stanford.edu/#faculty">Stanford’s Artificial Intelligence Laboratory</a> and more specifically to the lab of <a href="https://cs.stanford.edu/~chrismre/">Prof. Chris Ré</a>, a MacArthur genius grant recipient whose team has been producing cutting-edge research on a subset of machine learning techniques called “weak supervision.” The lab’s goal is to “make it faster and easier to inject what a human knows about the world into a machine learning model,” explains Alex Ratner, a PhD student who leads the lab’s open source weak supervision project, called Snorkel.</p><p>The predominant machine learning approach today is supervised learning, in which humans spend months or years hand-labeling millions of data points individually so computers can learn to predict events. For example, to train a machine learning model to predict whether a chest X-ray is abnormal or not, a radiologist might hand-label tens of thousands of radiographs as “normal” or “abnormal.”</p><p>The goal of Snorkel, and weak supervision techniques more broadly, is to let ‘domain experts’ (in our case, journalists) train machine learning models using functions or rules that automatically label data instead of the tedious and costly process of labeling by hand. Something along the lines of: “If you encounter problem x, tackle it this way.” (Here’s a <a href="https://hazyresearch.github.io/snorkel/">technical description</a> of Snorkel).</p><p>“We aim to democratize and speed up machine learning,” Ratner said when we first met last fall, which immediately got me thinking about the possible applications to investigative reporting. If Snorkel can help doctors quickly extract knowledge from troves of x-rays and CT scans to triage patients in a way that makes sense — rather than patients languishing in queue — it can likely also help journalists find leads and prioritize stories in Panama Papers-like situations.</p><p>Ratner also told me that he wasn’t interested in “needlessly fancy” solutions. He aims for the fastest and simplest way to solve each problem.</p><p>In early January, my newsroom, the <a href="https://www.icij.org/">International Consortium of Investigative Journalists</a>, and Re’s Stanford lab launched a collaboration that seeks to enhance the investigative reporting process. To honor the “nothing needlessly fancy” principle, we call it <em>Machine Learning for Investigations.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RMB66RbnDfgXIWEhtKaaOQ.jpeg" /><figcaption>ICIJ reporters and Stanford research scientists use machine learning in a journalistic investigation.</figcaption></figure><p>For journalists, the appeal of collaborating with academics is twofold: access to tools and techniques that can aid our reporting, and the absence of commercial purpose in the university setting. For academics, the appeal is the “real world” problems and datasets journalists bring to the table and, potentially, new technical challenges.</p><p>Here are lessons we learned so far in our partnership:</p><ul><li><strong>Pick an AI lab with “real world” applications track record. </strong>Chris Ré’s lab, for example, is part of a consortium of government and private sector organizations that developed a set of tools designed to “light up” the Dark Web. Using machine learning, law enforcement agencies were able to extract and visualize information — sometimes hidden inside images — that helped them <a href="https://www.scientificamerican.com/article/human-traffickers-caught-on-hidden-internet/">go after human trafficking networks</a> that thrive in the Internet. Searching the Panama Papers is not that different from searching the depths of the Dark Web. We have a lot to learn from the lab’s previous work.</li><li><strong>Make sure there are incentives on both sides. </strong>There are many civic-minded AI scientists concerned about the state of democracy who would like to help journalists do world-changing reporting. But for a partnership to last and be productive, it helps if there is a technical challenge academics can tackle, and if the data can be reproduced and published in an academic setting. Sort out early in the relationship if there’s goal alignment and what the trade-offs are. For us, it meant focusing first on a public data medical investigation because it fit well with research Ré’s lab was already doing to help doctors anticipate when a medical device might fail. The partnership is helping us build on the <a href="https://www.icij.org/investigations/implant-files/algorithms-analysis-and-adverse-events-how-icij-used-machine-learning-to-help-find-medical-device-issues/">machine learning work the ICIJ team did</a> last year for the award-winning <a href="https://www.icij.org/investigations/implant-files/">Implant Files investigation</a>, which exposed gross lack of regulation of medical devices worldwide.</li><li><strong>Choose useful not fancy.</strong> There are problems for which we don’t need machine learning at all. So how do we know when AI is the right choice? John Keefe, who leads Quartz AI Studio, says machine learning can help journalists in situations where they know what information they are looking for in large amounts of documents but finding it would take too long or would be too hard. Take the examples of Buzzfeed News’ 2017 <a href="https://www.buzzfeednews.com/article/peteraldhous/hidden-spy-planes">spy planes investigation</a> in which a machine learning algorithm was deployed on flight-tracking data to identify surveillance aircraft (here the computer had been taught the turning rates, speed and altitude patterns of spy planes), or the Atlanta Journal Constitution <a href="http://doctors.ajc.com/about_this_investigation/">probe on doctors’ sexual harassment</a>, in which a computer algorithm helped identify cases of sexual abuse in more than 100,000 disciplinary documents. I am also fascinated by the work of Ukrainian data journalism agency Texty, which used machine learning to <a href="http://texty.org.ua/d/2018/amber_eng/">uncover illegal sites of amber mining</a> through the analysis of 450,000 satellite images.</li><li><strong>“Reporter in the loop” all the way through.</strong> If you are using machine learning in your investigation, make sure to get buy in from reporters and editors involved in the project. You may find resistance because newsroom AI literacy is still quite low. At ICIJ, research editor Emilia Diaz-Struck has been the “AI translator” for our newsroom, helping journalists understand why and when we may choose to use machine learning. “The bottom line is that we use it to solve journalistic problems that otherwise wouldn’t get solved,” she says. Reporters play a big role in the AI process because they are the ‘domain experts’ that the computer needs to learn from — the equivalent to the radiologist who trains a model to recognize different levels of malignancy in a tumor. In the Implant Files investigation, reporters helped train a machine learning algorithm to systematically identify death reports that were <a href="https://www.icij.org/investigations/implant-files/how-medical-device-harm-is-concealed/">misclassified as injuries and malfunctions</a>, a trend first spotted by a source who tipped the journalists.</li><li><strong>It’s not magic!</strong> The computer is augmenting the work of a journalist not replacing it. The AJC team read all the documents associated to the more than 6,000 doctor sex abuse cases it found using machine learning. ICIJ fact-checkers manually reviewed each of the 2,100 deaths the algorithm uncovered. “The journalism doesn’t stop, it just gets a hop,” says Keefe. His team at Quartz recently received a <a href="https://qz.com/1464390/announcing-the-quartz-ai-studio-designed-to-help-journalists-use-machine-learning/">grant </a>from the Knight Foundation to partner with newsrooms on machine learning investigations.</li><li><strong>Share the experience so others can learn</strong>. In this area, journalists have much to learn from the academic tradition of building on one another’s knowledge and openly sharing results, both good and bad. “Failure is an important signal for researchers,” says Ratner. “When we work on a project that fails, as embarrassing as it is, that’s often what kicks off multiyear research projects. In these collaborations, failure is something that should be tracked and measured and reported.”</li></ul><p>So yes, you will be hearing from us either way!</p><p>There’s a ton of serendipity that can happen when two different worlds come together to tackle a problem. ICIJ’s data team has now started to collaborate with another part of Ré’s lab that specializes in extracting meaning and relationships from <a href="https://hazyresearch.github.io/snorkel/blog/fonduer.html">text that is “trapped” in tables</a> and other strange formats (think SEC documents or <a href="https://www.icij.org/investigations/luxembourg-leaks/your-head-spinning-5-tips-understand-lux-leaks-files/">head-spinning charts</a> from ICIJ’s Luxembourg Leaks project).</p><p>The lab is also working on other more futuristic applications, such as capturing natural language explanations from domain experts that can be used to train AI models (It’s appropriately called <a href="https://hazyresearch.github.io/snorkel/blog/babble_labble.html">Babble Labble</a>) or tracing radiologists’ eyes when they read a study to see if those signals can also help train algorithms.</p><p>Perhaps one day, not too far in the future, my ICIJ colleague Will Fitzgibbon will use Babble Labble to talk the computer’s ear off about his knowledge of money laundering. And we will trace my colleague Simon Bowers’ eyes when he interprets those impossible, multi-step charts that, when unlocked, <a href="http://www.tedxglasgow.com/2018/04/simon-bowers-talks-tax-at-tedxglasgow-2018/">reveal the schemes</a> multinational companies use to avoid paying taxes.</p><p>In the meantime, we stay real. Nothing needlessly fancy.</p><p><strong><em>Are you a journalist with a story idea or data that would benefit from machine learning? Are you a machine learning expert interested in collaborating with journalists? Reach out to me and let’s chat about ways to collaborate: </em></strong><a href="/service/mailto:marinawg@stanford.edu"><strong><em>marinawg@stanford.edu</em></strong></a><strong><em> @MarinaWalkerG</em></strong></p><p>The following people are part of the Machine Learning for Investigations partnership between ICIJ and Stanford: Alex Ratner, Jason Fries, Jared Dunnmon, Mandy Lu, Alison Callahan, Emilia Diaz-Struck, Rigoberto Carvajal, Sen Wu.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1f97164742a0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jsk-class-of-2019/we-cracked-the-panama-papers-with-400-human-brains-can-ai-help-us-next-time-1f97164742a0">We cracked the Panama Papers with 400 human brains. Can AI help us next time?</a> was originally published in <a href="https://medium.com/jsk-class-of-2019">JSK Class of 2019</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I came to Stanford to study artificial intelligence and decided to tackle my brain first]]></title>
            <link>https://medium.com/jsk-class-of-2019/i-came-to-stanford-to-study-artificial-intelligence-and-decided-to-tackle-my-brain-first-c44897ff563b?source=rss-184b8833d2fe------2</link>
            <guid isPermaLink="false">https://medium.com/p/c44897ff563b</guid>
            <category><![CDATA[self-reflection]]></category>
            <category><![CDATA[digital-media]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[brain-training]]></category>
            <category><![CDATA[journalism]]></category>
            <dc:creator><![CDATA[Marina Walker Guevara]]></dc:creator>
            <pubDate>Thu, 06 Dec 2018 20:15:22 GMT</pubDate>
            <atom:updated>2018-12-06T20:15:22.640Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Z12ZivwL3MA6rkOOKF_OZg.jpeg" /><figcaption>Credit: <a href="https://www.flickr.com/photos/olfiika/8744511730/in/photolist-ejHUFC-bPSHpn-hpyfCw-YnjLin-jg7tJB-i2TDMK-2bKauYg-XFtZS7-8tatTQ-scyJrC-bA2THe-qhAgU-cqbH69-ih8gDL-75mpDC-uYr5KL-7kPgHf-m3ZR64-5FRKb4-21dQN1v-cXjhC5-cXjhH5-b8x1Ca-g9s4X4-o7zsN-fLov3g-894j54-6SLGcW-4D3HSR-4epgz8-fLELAA-9cLtX3-CQNPyB-pMS2se-ata9Ar-cXjqJj-cXjkKj-6FMas5-fLENAE-dMDfJ2-fLEZc3-bEdPrY-bA2VAg-nBEDiD-eHowQ5-eHhrW4-c2yQA3-bntmhT-6w9fWS-6UhVrV">Olfiika</a></figcaption></figure><p><strong>The year I set out to interrogate higher forms of computer intelligence to chase bad guys, the first lesson was a deeply personal one.</strong></p><p>“Hep!” the instructor yelled.</p><p>I jumped from the platform, 22 feet above the ground, and held tight with both hands to the trapeze bar: a true “leap of faith” for someone who had never been jogging, much less tried an extreme sport. At the instructor’s command, I pulled my legs up, hooked them over the bar as the trapeze swung forward, then let my hands go. In that upside-down position, I arched my back and extended my arms forward so, if the plan worked, the professional trapeze partner — swinging on the other end of the arena — could catch me in the air. We locked arms in the middle, and in the biggest act of trust yet, I let go of my trapeze. (I was in a harness but still…)</p><p>For years, I had used a trapeze analogy in many training sessions I’d conducted around the world about collaborative investigative reporting — the idea of shared ownership of a story in which we all take the plunge knowing that, ultimately, we have each other’s backs.</p><p>So when my Stanford professors at the d.school proposed a series of “self-discovery” missions for our Transformative Design class, I did not hesitate to go for the trapeze option.</p><p>I came to Stanford in September on a John S. Knight Journalism Fellowship to study how we might <a href="https://jsk.stanford.edu/fellows/class-of-2019/marina-walker-guevara/">democratize artificial intelligence among investigative reporters</a>, so we can unravel corruption more efficiently. I had planned to spend most of my time from day one on this exciting project so the journalists in my network — the <a href="https://www.icij.org/">International Consortium of Investigative Journalists</a> — are better equipped to tackle the next <a href="https://www.icij.org/investigations/panama-papers/">Panama Papers</a> when it arrives.</p><p>Then the trapeze assignment disrupted my priorities, at least temporarily.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*CYMjE3B4RQ7TBC8tXn5sSA.gif" /></figure><p>“Pushing boundaries,” I wrote on Facebook as I shared the video of my Wallenda-like feat. Upon more reflection, I realized that I wasn’t being brave or audacious at all.</p><p>Just like computer algorithms learn what we humans teach them and their outputs reflect our biases and those of the data they ingest, my trapeze challenge was a perfect manifestation of my biases — and an escape from what I really needed to be working on.</p><p>In short, I told my classmates, I am okay with the big jumps. That’s not my issue. I thrive on big challenges. The same roller-coaster feeling I had when I jumped off that high board, I experienced when I managed the Panama Papers and other complex global investigations. And when I co-led the breakaway of my organization, ICIJ, from its parent company. And, on a more personal level, when at 17, I wrestled anorexia to the ground and sought the help that ultimately saved my life.</p><p>I am brave on the big stuff. It’s “little” stuff that I have tended to ignore. It was no wonder, then, that after many years of journalistic thrills at my job — where we truly felt we could change the world — other parts of life had become numb, like when receptors in our brains don’t get the necessary signals to trigger a response.</p><p>All things considered, I thought I had maintained a good “balance” as a mother of two, a partner, a friend and a busy journalist. I had seen colleagues burn out right and left — and I still do — but always considered myself safe from those travails. I was even a bit impatient with their struggle to make it all work.</p><p>That was until I came to Stanford for my year-long fellowship, leaving behind in Washington D.C. that bubble called work. The absence of work, with all the identity and validation I attach to it, revealed deserts in my life that I had not dared to admit: exercising, reading for pleasure, spending time alone, praying, dancing, volunteering and being a better partner to my husband.</p><p>I felt a California-size earthquake under my feet for weeks on end.</p><p>So, in the year I set out to interrogate high forms of computer intelligence to better chase bad guys, I decided to start by retraining my own brain to become a healthier, better version of myself.</p><p>In the d.school class, I was paired with Dakota, a quiet sophomore from Ohio. The assignment: Design behaviors that addressed issues in each other’s lives. It didn’t take her long to realize that the most game-changing behavior for my life, as she put it, would be to learn to do nothing.</p><p>I had mentioned to her that I usually have a cup of tea in the middle of the morning and afternoon while reading news, catching up with social media or working. Productive stuff. So, she instructed me to use the cup of tea as a trigger but to remove all other distractions. Good designers work with what’s already there in our lives and give it a transformative twist, rather than introduce a bright new object with no connection to our experience. She called the exercise Sip &amp; Savor and sent me on my way.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*13zIYG3Jbq3gmFPWozATHA.jpeg" /><figcaption>A 100 Post-It brainstorm for new, healthy behaviors.</figcaption></figure><p>I failed at many of my early attempts. As I sat on my sun-filled Palo Alto doorstep or on campus holding my cup of tea, I would catch myself doing homework or making to-do lists in my head. Or formulating Halloween plans.</p><p>Just like with machine learning, repetition helps. As I engaged more and more in this new behavior, my brain started to get it and to go along with it. My senses opened. I discovered the smell of the eucalyptus trees on campus and spotted for the first time the tiny labels our landlords have sprinkled throughout our front yard, identifying each plant by its Latin name. As I paid attention to my breathing, I remembered with gratitude how learning to send air to my stomach, rather than my chest, rescued me from many asthma attacks as a child.</p><p>My cousin from Argentina, also named Marina, captured the essence of my first quarter at Stanford better than I can explain:</p><p><em>We must learn to find beauty in the simplicity of making a cup of tea. When in those small acts we find beauty, peace and harmony, life acquires new meaning. … It produces a change, a big leap inside us and toward humanity. Keep making cups of tea.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c44897ff563b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jsk-class-of-2019/i-came-to-stanford-to-study-artificial-intelligence-and-decided-to-tackle-my-brain-first-c44897ff563b">I came to Stanford to study artificial intelligence and decided to tackle my brain first</a> was originally published in <a href="https://medium.com/jsk-class-of-2019">JSK Class of 2019</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>