Solving ‘Fake News’ with AI, Blockchain and a Global Community

An immutable registry of labeled ‘fake news’ and other classes of misleading content, accessible by all — humans and machines — to help quantify the problem and raise global awareness.

Published in

The Innovation Machine

5 min readJul 26, 2019

Fake news is a major problem in the era of AI. Although, similar dysfunctional scenarios involving misinformation and propaganda have been around for ages — in one form or another — the problem of ‘fake news’ is becoming a real threat to our connected society. It is the ease of creating, diffusing, and circulating such content, combined with the difficulty to track it and control it on time, which makes it a significant threat and a hard-to-solve problem.

Especially with the so-called ‘deepfakes’, we will soon be unable to tell what is true and what is not: the times when something was perceived as true just because it was ‘seen on TV’ or in a video, are gone. We have entered a period of a ‘high-tech propaganda’ in which people and media are found to participate at scale — intentionally or not.

The ‘Viral Nature’ of ‘Fake News’

Fake content is designed to be viral; its creators want it to spread organically and rapidly. Fake stories are engineered to attract attention and trigger emotional reactions so users instantly share them with their networks. With the right tricks and timing, a false story can go viral in hours. The ‘fake news industry’ takes advantage of the following ‘flaws’ or our online reality:

1. The online network is optimized for ‘click-through’

The global ‘news distribution network’, including social media, news corporations, opinion leaders and influencers is typically optimized for clicks. Quality and trustworthiness of content typically come next. As a result, websites and other online entities rush to reproduce stories that appear to be potentially viral; and they distribute them; so they get more traffic and serve more ads, to achieve their ambitious monetization goals. Content Quality is usually not part of their KPIs — goals are based on CTRs, page views, and related metrics; and when there are complaints about poor content, they simply remove it.

2. Online users are ‘optimized’ for ‘instant sharing’

There is another layer on top of this ‘news distribution network’: the online users — who are optimized for sharing. It is sad to realize that in an era of unprecedented access to world’s knowledge, the majority of the online users are ‘passive re-sharers’; they don’t create original content; they just recycle whatever appears to be ‘cool’ or trendy, with little or no judgment and critical thinking. Users of this class, consume and circulate fake news — and other poor content — and they become key components of the mechanism. Unintentionally.

A Problem of Awareness

Obviously, there are entities who intentionally drive fake news (to achieve certain goals) and others who unintentionally participate in the exponential spread across the globe. The latter — being individuals or companies — will probably never realize that they are part of the fake news problem. A significant component of a great solution is awareness and quantification. And this can be achieved by using the latest and greatest technological achievements: Artificial Intelligence and Blockchain.

There are significant efforts within corporations and social media companies to mitigate the problem. And some of them may prove to be effective. But the fake news problem is bigger — it goes beyond the corporate boundaries. It requires a global collaboration network, shared resources, and knowledge; a powerful shared framework. We need a global registry of fake news and misleading content. And a global community to keep it up to date.

We need a way to continually identify and label ‘Fake News’ — both to measure their impact and also to learn from the patterns.

Fake news can be identified — with human effort supported by AI/ advanced NLP/ language understanding and pattern detection. The problem is that it is difficult to do so early enough — before a fake story starts shaping the public opinion or triggering rumors and false impressions.

We could take a different approach and focus on quantifying the level of responsibility of each of the involved parties with the intent not to penalize but to educate and raise global awareness.

Consider a global, immutable registry of labeled ‘fake news’ and other classes of misleading content. A ‘fake news’ registry, accessible by all — humans and machines. By all media and news companies in the world. This could be blockchain pointing to unified content stored in IPFS or Swarm — an immutable system hosting samples of the world’s content, unified and labeled in terms of trustworthiness.

Imagine the following process running on a daily basis — sampling the global content publishing and sharing activity.

Crawlers discover ‘fresh content’ and ‘new content references’. Special crawlers — similar to those used by search engines — run continuously across a representative set of major news sites, social media, popular blogs, feeds, news organizations. They have two objectives: [a] to identify ‘fresh content’ — the delta published in a specific time frame [b] to identify new listings of the already registered, unified content.
Crawlers post the ‘fresh content’ and references into an immutable data store. The fresh content identified, is unified and linked to its ‘master copy’. References of the instance of the master copy are appended — also pointing to the website and the entity. Unique identifiers of the content or of the posts sharing the content are also appended into the immutable store.
AI + Humans spot problematic content. A global community of professionals and digital citizens is discovering, labeling, and justifying problematic content. With the assistance from advanced AI components making suggestions for review of suspicious content and automatically linking ‘versions of the same content’, this global community labels the problematic content.
Media entities self-assess their compliance and progress ‘towards better content for the world’. When a ‘fake news label’ is confirmed for a piece of content (also with a classification of the issue) it joins the ‘media entity evaluation system’ — a scoring system quantifying how website A or social media B or news corporation C is part of the global fake news problem (or within a category). Having this information, media entities can take action, learn and measure how responsible they are in spreading the particular fake story (and all the other labeled content).

They can let their users know that certain stories they have shared, proved to be false and misleading. They can help the global effort by educating their users and demonstrating responsibility and real actions towards a better-informed society. Social media and new companies could integrate calls to the APIs of this global registry and cross-check content at ‘share time’ and notify the user if the content is flagged. Or notify users who have already engaged (liked, shared, saved, commented on) with ‘verified fake news’ stories.

There are countless interesting use cases — including proper quantification, global trend analysis, and articulation of the dynamics of the phenomenon. A global effort on top of an immutable registry of unified labeled fake news content. Powered by humans and A.I.

References