How newsrooms fight disinformation – a Q&A with two experts

How newsrooms fight disinformation – a Q&A with two experts

 

How newsrooms fight disinformation
Q&A with Julia Bayer and Ruben Bouwmeester, DW Research and Cooperation projects

Tuesday, December 14 • 12:00-13:00 CET • via Zoom

How do journalists work against mis- and disinformation? What are the workflows for the verification of content? What are the main challenges for newsrooms? And: What do the experts expect from future blockchain solutions? To get answers and insights please join us for an open webinar on December 14, 2021, from 12:00 – 13:00. 

Join us as a guest

The event is organised for the now 25 funded project teams of TruBlo. But participation is open for guests from the NGI community, EU researchers, startups and developers. If you want to join as a guest please contact us at support@trublo.eu and we sent you the log-in credentials for the Zoom conference. 

The experts will give a very brief overview, the session will then allow for Q&A by the participants.

What we discuss:

  • How do journalists cope with disinformation?
  • What are the workflows?
  • Is there a way to overcome the current crisis of trust?
  • What tools are available and working?
  • What are current views on blockchain technology used for trustable content?

As part of the talk, the experts will talk about tools created to fight disinformation and to help with verification. Examples are the InVid verification plug-in, work done in the EU research project Digger (Deepfake detection) and Truly Media. The latter is a software platform for collaborative verification. Truly Media is used by a number of organisations, e.g. Amnesty International, Reuters, ZDF and the European Parliament.

Link: https://www.truly.media

Our experts:

Julia Bayer is an investigative journalist who dives deep online using OSINT. She has gained first-hand experience from work in the social media team of Deutsche Welle. In addition, she works as an innovation manager of the Research- and Cooperation team of Deutsche Welle. Julia is involved in a number of verification projects, such as the implementation of Truly Media and InVid. She also trains journalists in OSINT techniques and digital verification. Julia has worked with journalists in multiple international locations and has worked with newsrooms, NGOs and DW academy. Julia is the founder of @quiztime on Twitter, where a worldwide community of participants regularly has to solve verification and geolocation tasks.

Twitter: @bayer_julia

Ruben Bouwmeester brings concept development skills and profound graphical expertise. In his role as an innovation manager, Ruben was a key contributor to several software solutions and platforms used for the detection of disinformation and deep fakes. Ruben has been a key architect and manager for Truly Media, a collaboration platform created by the DW ReCo team in collaboration with Athens Technology Center. He has a background in project management, concept development, and design, holding a B.A. from The Hague University. As a ReCo innovation manager, Ruben specialises in User Generated Content (UGC), verification, and HLT. He also trained journalists at DW Akademie.

Twitter: @rubybouw

We need to rethink what we consider as “content”

We need to rethink what we consider as “content”

The fight against misinformation can be compared to a big clean-up initiative: Think of your neighbourhood is covered with trash, not only from last week but several years. Littering everything is done quickly, but cleaning up takes much longer.

What is important here is to rethink, critically, what we consider as content. Yes, a news article is clearly content. But there are today many other forms of information bits that are either preceding or following pieces of traditional news media content. And it should be clear that all these added bits of information must be reliable and trustable. 

When falsified content goes viral, lives might be in danger

One big criticism towards large social media platforms is about not having foreseen and then later not having acted against dynamically generated information, based on posts, comments, discussions. When falsified content goes viral, lives can be in danger. There are documented cases where false information was used to incite anger among a group, sometimes leading to angry mobs in the streets burning down the house of a victim of such allegations

We must, as a result, be clear that all elements of what we consider trustable information must be verifiable. Content components include of course written text. But the definition of what is the content must include pictures, artwork/visuals, videos, video stills, numerical data, system data, raw or aggregated data, algorithms. In addition, we must include user-generated content, such as discussions, opinions and other interactions on social media. So far, on a technical level, it is very difficult to separate one from the other, if only the words are analysed. One take-away is that all kind of meta-data must be included in the analysis, too. 

Most difficult: Mixing true and false information

A very difficult problem here is when correct content is taken out of context and is combined or mixed with false or fabricated information – the connection is hard to distinguish, specifically by automated searches. Humans might be able to see the difference, but there is no feasible way that every item of information is checked for plausibility or truth by a human. Search strings and search methods to identify new information are as important to evaluate as other forms of information detection. 

Just one example to illustrate how technical systems can be tricked is the manipulation of publication dates for information. A malicious actor might have written false content, with catchy headlines. In order to let a search spider pick up the content as new, it is sufficient to re-publish the content, potentially on a different website and under a different URL and IP address. 

Fraud detection can be tricked

Much of the data used by Google for search depends on what website owners and content creators provide. Of course, there are all kinds of fraud detection, but in a world where not many data points for content that is published via Content Management Systems can be verified at the source, the search engines do have not many options. Further, of course, the number of content sources that are intentionally falsifying what they publish is always only a fraction of the total. But because they have a chance to go undetected they can do so much harm. 

If a website methodically refreshes dates for the content on its pages there are not many ways for search engines to detect this. Or, in other words: With the right motivation and a little bit of know-how, it is possible to trick search spider software into believing that a recycled article has been published very recently.

What is the main motivation to invest work into falsified content? Presumably, the main and the most frequent motivation is simply to make money. Running a partially automated fake news system and connecting it to an advertising platform can result in a very good payout. 

Perspective: $50 billion lost to ad fraud by 2025

A report published by industry organisation IAB Europe says: “According to the World Federation of Advertisers (WFA), it is estimated that by 2025, over $50 billion will be wasted annually on ad fraud.” 

After the 2016 election researchers found that a considerable number of entirely faked articles were coming from a region as far away from the US as Macedonia. Some people there had learned how to make money through digital advertising and the key was to write entirely falsified, but outrageous articles about political candidates. The wilder the allegations, the better the click rates and shares for such content. This created a mini-industry based on “fake news” in the region. 

The techniques which make ad fraud successful can be used for political or criminal disinformation campaigns. The financial motivation for ad fraud exploits helps to build experience and a lot of practical knowledge on how to mislead existing platforms, which then can be reused for targeted disinformation campaigns. 

These are some, but not even all arguments why we need to broaden our understanding of what is content. Over time there should be detection measures, even at the source where the information is published, to enable 100% verification. 

Examples of funded projects from TruBlo

Among the ten projects which received funding in the 1st open call are several which are explicitly looking for new ways to detect falsified information and content. Some examples below, full list can be found here.

CONTOUR – Trusted content for tourism marketing purposes

LEDGEAIR – Aircraft data mining framework

ShoppEx – to restore the trust between retailers/brands and consumers

More information:

IAB Europe: Guide to ad fraud, 2020

 

 

To fight misinformation,  the “Coalition for Content Provenance and Authenticity” (C2PA) wants to develop technical standards

To fight misinformation, the “Coalition for Content Provenance and Authenticity” (C2PA) wants to develop technical standards

In February 2021 the “Coalition for Content Provenance and Authenticity” (C2PA) combined and consolidated the efforts of two projects with similar goals. On one side “Project Origin”, founded in 2020 by Microsoft, BBC, The New York Times and CBC/Radio-Canada.
On the other side a group by the name of “Content Authenticity Initiative”, founded by Adobe. In the new, larger group more members are represented, including the chip-makers Arm and Intel. TruePic, a startup with interesting technology in this space is a member, too.
What all the participants have in common is this: To develop standards and tools for reliable content provenance. This would include certifying the source, the origin and the history of content elements. But, it is not an attempt to re-create DRM (Digital Rights Management).

Today falsifying content is easy

So far, content that is accessible on the internet can be intentionally falsified, easily.

  • Tricking search engines: From false claims to falsified sources to sloppy or falsified metadata. For example a low level, but common approach to content fraud is to simply change the publishing date. Search engines are good at finding content, but all search platforms are challenged by falsified data and information.
  • No restrictions in Content Management Systems: For the sake of convenience, almost all currently used Content Management Systems do not impose strict guidelines nor checks for copyrights or whether the publishing data is correct.

In earlier communication the group behind Project Origin stated the goal: “Having a provable source of origin for media, and knowing that the content had not tampered with en-route, will help to maintain confidence in news from trusted providers”.

Technical demonstrators so far

How do they want to get there? So far the concepts are demonstrators or software in beta. Technically, the idea is to define an “end-to-end process for the publishing, distribution and presentation of provenance enhanced media”. Provenance is of course the key here. It means that added information should enable to trace down where the content came from and whether what is presented is actually the version that was published at the origin. Provenance information will be added no only to text-based media but also to audio, video and images.

In September 2020 the project origin published a short video, which narrated the motivation as well as the early proof-of-concept models. Key goals are to confirm the identity of the publisher and to ensure that the content has not been tampered with. This applies to the visible editorial parts, but also to metadata which might not be directly visible, but is used for example by search engines to rank the content. It is common for misinformation to change dates, for example, to make old content look “new” again.

Two videos, one from Project Origin, the other from the Adobe-led Content Authenticity Initiative show the intentions and goals.

Project Origin (2020)

Source: BBC

Content Authenticity Initiative: Vision

Searching for a standard

As of spring 2021, there is a common goal and specific roles for the different partners.  “Coalition for Content Provenance and Authenticity” (C2PA) aims to advocate and find support for an open standard. At the same time the different technology companies work on their individual solutions:

  • Adobe is working on an attribution tool, which could be added to Photoshop and other Adobe packages.
  • TruePic, a start-up from San Diego in the US, could have the most innovative. The company has developed software specifically to enhance and enrich content enabling checks of its integrity.
  • Microsoft describes the current approaches on a dedicated innovation website.
  • It is notable that Google is missing from the members, so far. The same is true for Apple and Facebook.
    It might be that if all the big tech and content companies would join finding a compromise would become almost impossible. Instead, that is the assumption, the smaller, current group firstly wants to develop early suggestions and ideas.

Demos from Microsoft

Microsoft specifically has the most details yet how Project Origin could work. The company discusses current ideas on a webpage of the innovation department. They christened the technical approach AMP (Authentication of Media via Provenance).

Search for standard might take years

  • Agreeing on standards, specifically, might take years. While misinformation and content used for propaganda in many parts of the world are pressing problems. Finding common ground will depend on many details – and take time, specifically when there is a need for an agreed-upon technical standard, which is accepted worldwide.
  • One thing that is notable: All three projects and the entire coalition do not commit to a specific blockchain technology to store the information. It appears that the question is currently avoided – there might or might not blockchain tech in the backend.
  • Another problem is to find a technology that gets accepted by users and does not make media production more complicated. Adding metadata is a notoriously skipped activity in many fields of content production.

More info:

Microsoft Innovation: Exploring Project Origin
https://innovation.microsoft.com/en-us/exploring-project-origin

Microsoft Innovation: Technical explanation and demo
Deep Dive: Technical explanation of Project AMP’s components including a demo based on the paper
AMP: Authentication of Media via Provenance.

Adobe brings its misinformation-fighting content attribution tool to the Photoshop beta
https://techcrunch.com/2020/10/20/adobe-cai-photoshop-beta-misinformation/

Newsletter #11: TruBlo project update • Carrots, not sticks to quell misinformation and more…

Newsletter #11: TruBlo project update • Carrots, not sticks to quell misinformation and more…

TruBlo Project Update

Next steps for Trublo open call #1: If you participated, when will you get results? We published an article on the TruBlo website, to provide an overview of the process. The quick answer to the key question: Participants can expect a notification on April 30, 2021. More details: [TruBlo Website](https://www.trublo.eu/2021/04/09/trublo-open-call-1-next-steps-and-selection-process-overview/)

Updates this week:


TRUST


Trust in tech declining, survey finds

 ”Trust in tech — including companies specializing in AI, VR, 5G and the internet of things — fell all around the world last year, the Edelman Trust Barometer found in a massive survey of 31,000 people in 27 countries.”
[LINK


Procter & Gamble accused of collaboration with Chinese advertising

Apple aims to change how data can be collected for advertising purposes, specifically how profiles of users can be created. Many advertisers and other platforms are opposing the change. One unexpected result: The US company Procter & Gamble teaming up trade groups in China.

According to a report from the Wall Street Journal:

”Procter & Gamble Co. helped develop a technique being tested in China to gather iPhone data for targeted ads, a step intended to give companies a way around Apple Inc.’s new privacy tools, according to people familiar with the matter. […]
The company has joined forces with dozens of Chinese trade groups and tech firms working with the state-backed China Advertising Association to develop the new technique, which would use a technology called device fingerprinting.” LINK


Blockchain for trustable food system – from seafood to grain

The Conversation: “With global-scale food systems such as seafood, nearly 40 per cent of which is traded globally, data transparency and traceability through technologies like blockchain are important for socially and environmentally conscious decision making and to facilitate trust among stakeholders. Blockchain technologies can be used to consolidate information on the quality of the seed, track how crops grow and record the journey once it leaves the farm.” LINK


Content


New motivation: ”To quell misinformation, use carrots – not just sticks”

An article by neuroscientist Tali Sharot published in “Nature” argues for a fundamental change in how users should be rewarded when posting content.

Quote: “Most readers have felt an ego boost when their post received ‘likes’. Such engagement also results in followers, which can help people secure lucrative deals. Thus, if a certain type of content generates high engagement, people will post more content like it. Here is the conundrum: fake news generates more retweets and likes than do reliable posts, spreading 6–20 times faster.”

What could be a solution?

”At the moment, users are rewarded when their post appeals to the masses — even if it’s of poor quality. What would happen if users were rewarded for reliability and accuracy?” LINK


Facebook criticised for being unresponsive to reports of manipulative content, investigation reveals

The Guardian published findings of an investigation of how the social platform handled manipulative content, specifically outside of the US. A former employee says that Facebook was often inactive, even when warned about such manipulative content.

“There is a lot of harm being done on Facebook that is not being responded to because it is not considered enough of a PR risk to Facebook,” said Sophie Zhang, a former data scientist at Facebook who worked within the company’s “integrity” organization to combat inauthentic behaviour. “The cost isn’t borne by Facebook. It’s borne by the broader world as a whole.” LINK


BLOCKCHAIN


Revenge of the Winklevii

Forbes reports at length about how Cameron and Tyler Winklevoss are emerging as investors in several blockchain and crypto projects.

A dozen years after they settled with Zuckerberg for $65 million in Facebook stock and cash, the Winklevii, as they are widely known, have emerged as leaders of a technological movement whose core operating principle involves digitizing the records of all assets globally, decentralizing control and cutting out gatekeepers—including Facebook.

The two are now owners of a holding company called “Gemini Space Station”, which “owns their crypto exchange and Nifty Gateway”. Forbes reports that the twins invested in 25 digital asset startups. LINK


A notable uptick of interest in cryptocurrencies and blockchain initiatives

The rise of Bitcoin value is driving interest in trading of cryptocurrency trading. In parallel, we see new corporate projects for blockchain. While the two areas are not directly related, the financial market seems to validate blockchain technology, to an extend.

Selected examples:

  • Trading app Robinhood says 9.5 million users traded crypto in Q1 2021. This is an increase of 458% related to just 1,7 million traders in Q4/2020. LINK
  • The Wall Street Journal has a special how “GameStop, Blockchain.com and Bitcoin Renewed a Push to Digitize the Stock Market”. A lot of speculation mixed with enthusiasm, along with potential volatility and other risks. LINK
  • This week it’s the second virtual edition of the European Blockchain Convention, due to the pandemic. It is worth your time to scroll through the (long) list of attendees – you’ll find a mix of blockchain start-ups, representatives from larger companies and many EU officials. Just the visual scroll will give you an idea of interest in all things blockchain. LINK
  • And finally: Drivers of the new Fiat 500 electric vehicle will earn coins if they drive sustainably:

”Stellantis and UK-based startup Kiri Technologies will use blockchain rewards to encourage sustainable driving behaviour. Stellantis is the result of a recent merger between Groupe PSA and Fiat Chrysler and owns 16 car brands, including Peugeot, Citroen, Chrysler and Fiat. The rewards initiative will be conducted under Stellantis e-mobility program. Drivers of the New 500 Fiat, a fully electric car, will be awarded KiriCoins, which can be spent in the Kiri marketplace.”


Thank You for reading.

Please forward this newsletter if you think it has value for a colleague or friend.

Do you have feedback or suggestions? [Contact us](mailto:info@trublo.eu)


Online conference: How can we fight disinformation using data and technology?

Online conference: How can we fight disinformation using data and technology?

Organized by Fandango, an EU co-funded project, the event will focus on new approaches as well as the role of the EU ICT policies in this field.
How can we create new tools? How can AI be used to detect misinformation? Is there a way to expose falsified or misleading content?
On the agenda:
  • The role of the EU’s ICT policies to protect European values, markets and citizens against disinformation.
  • Rethinking the possibilities of data and technology in the fight against disinformation.
  • EU’s ICT ecosystem against disinformation: the challenges ahead

This is the full agenda.

The discussion is open to everybody. Feel free to share the invitation with others who might be interested.
When? 
March 26th, Friday at 09:00 am (CET)

The Study About Dis-Information You Should Read

“Information Disorder” is one of the best studies to read if you want to get involved in the fight for trustable content. Written by Claire Wardle and Hossein Derakshan, two experts in the field, the study provides a structured overview. This helps to first understand the problem of false and falsified information. The study goes beyond simply describing the problem.  Instead, it provides a “framework for policy-makers, legislators, researchers, technologists and practitioners” who need to work together for results. 

 

A need for collaboration

 

The report acknowledges that the phenomena are not new. False rumours and campaigns using false information existed before the internet. What has changed recently is the scale of “information pollution”. False information can spread faster, making it more complex to provide correct facts. “Information Disorder” suggests that there is an urgent need to work collaboratively to find solutions. To get there the report provides a detailed framework of possible actions. 

 

As a helpful overview, the report visualises the three key forms of information disorder, side by side. 

 

Information disorder

From just false to outright harmful: Forms of information disorder. Source: Information Disorder, 2017.

  • Mis-information is when false information is shared, but no harm is meant.
  • Dis-information is when false information is knowingly shared to cause harm.
  • Mal-information is when genuine information is shared to cause harm, often by moving information designed to stay private in the public sphere.

The study suggests clearly to not use the label “fake news”, with reason:

“In this report, we refrain from using the term ‘fake news’, for two reasons. First, it is woefully inadequate to describe the complex phenomena of information pollution. The term has also begun to be appropriated by politicians around the world to describe news organisations whose coverage they find disagreeable. In this way, it’s becoming a mechanism by which the powerful can clamp down upon, restrict, undermine and circumvent the free press.”

Information Disorder, 2017

 

Why it matters: If we want to curb the amount of wrong information we must understand first that an unintended communication mistake has to be separated from a campaign using entirely fabricated info. This applies to both technical tools against such information as well other forms of activism in this space.

The study asks the key questions: Who is creating disinformation? Why are they doing it and what are their goals? This is extremely important for any initiative in this field: Only when we understand why some groups use this instrument we have a real chance to block, filter or expose intentional mis- and mal-information. 

As the study comes from two authors with a writing/journalism background they take a look at how and why people consume such content and why – at times – they start to belief in what is presented them and even re-distribute the content. Why is this? “A key argument within this report, which draws from the work of the scholar James Carey, is that we need to understand the ritualistic function of communication. Rather than simply thinking about communication as the transmission of information from one person to another, we must recognize that communication plays a fundamental role in representing shared beliefs. It is not just information, but drama”

The role of social platforms: “Information pollution” can affect all complex topics – health and medical, economy and ecology. As such the occurrence of false information and the use as a tool is not new. But a difference now is that social media platforms amplify the reach and encourage users to post material which might earn them approval from others – through likes and other subtle forms of positive feedback.

“It is not just information, but drama”

INFORMATION DISORDER, 2017

Step by step the authors develop a framework of recommendations and actions for multiple interest groups: Administrations, public bodies, governments, the civil society and – of course – media organisations. What makes the study helpful is that these lists of tips are straight-forward and simple. In combination, one could hope though that the numerous counter-activities help to fight information disorder, over time.

 

Download the 2018 version of the report here: Information Disorder (free, no registration)