DeepTruth.ai

The proliferation of “artificial reality” has become a permanent aspect of our browsing experience, specifically on social networks. The ability to create ultra-realistic “deep fake” videos using artificial intelligence means that no form of digital content can be trusted to be an accurate portrayal of reality anymore.

To counter this rising trend of false information, the crowdsourcing of truth through a blockchain-based decentralized system will leverage artificial intelligence to complement the wisdom of the crowd. This is where DeepTruth is introduced.

The DeepTruth decentralized web browser and platform will offer its users an array of algorithms specialized in detecting deep fakes. 

With the scaling factor of DeepTruth as a browser, millions, if not billions of people can partake in defining the truth in our ever growing digital world.

In short, DeepTruth will provide Truth-as-a-Service (TaaS)

The proliferation of “artificial reality” has become a permanent aspect of our browsing experience, specifically on social networks. The ability to create ultra-realistic “deep fake” videos using artificial intelligence means that no form of digital content can be trusted to be an accurate portrayal of reality anymore.

The tech platforms that were built to capture our attention are now being pressured to take on the responsibility of identifying and removing false information from their ecosystems. Even if the tech companies decide to become the arbiters of truth, will they be able to fulfill their responsibilities in a transparent, unbiased, and effective manner? Why should we trust those that own and control these tech platforms with safeguarding our shared truths?

By creating an immutable historical record of all of the validated to be false information, detailing where it originated from, who made changes to it, where it was shared, the DeepTruth DApp will create a “truth provenance” in its blockchain layer which will not only help to refine DeepTruth AIs, but will also create a valuable competitive advantage. 

Amongst other things, those individuals and businesses that are concerned about the impact of false information, like deep fakes, will be able to utilize our services for their timely detection and reporting.

As a decentralized web browser and platform, we are able to learn and work with those who are affected by deep fake media each and every minute of the day. We will be able to adjust our product in an expedited fasion to continue to meet our users needs.

DeepTruth will enable users and the world by offering a platform of truth that is defined by a blockchain based combination of Human and Artificial Intelligence.

DeepTruth serves as a decentralized web browser, extension, and app that will integrate a hybrid system composed human labor and artificial intelligence in order to detect online falsehoods in real time. 

Users will be able to shape their DeepTruth interface to either block, highlight, decolorize, or label any information that is judged to be inauthentic. 

Blockchain technology will allow the DeepTruth system to autonomously reward those users who act as truth “validators” and falsehood “seekers”. In parallel, DeepTruth algorithms will continuously learn from the consensus reached by “validators” and “seekers” of the crowd in order to reach higher levels of accuracy of detecting fabricated digital content.

DeepTruth leverages AI and blockchain to offer Truth-as-a-Service (TaaS).

The DeepTruth decentralized application will offer its users an array of algorithms specialized in detecting deep fakes, as well as to verify the accuracy of online information. 

Utilizing AI allows the DeepTruth browser to interoperate, creating a synergistic ecosystem capable of serving a more comprehensive set of tasks. 

There are four layers within DeepTruth.ai, symbolised in our logo by the circle, triangle, square, and sphere. These four layers include Blockchain, AI, UI/UX, and Truth Emerging. AI models that are trained to fight disinformation will be the products of a diverse variety of datasets that no single corporation can create on their own.

As deep fakes become incredibly realistic and easy to produce, their impact will be qualitatively different from what we have seen before. In the near future, we see the DeepTruth system forming a layer of authenticity that will become just as necessary for browsing the web as anti-viruses or ad-blockers are today.

We are in the process of filing a provisional patent for the DeepTruth Browser and it's technological capabilities.

Artificial and Human Intelligence Hybridization

Blockchain Technology

We believe the crowdsourcing of truth should be as democratic, decentralized, and transparent as possible. Rather than relying on systems that utilize only AI or localized human workforces to fight false information, DeepTruth incentivizes diverse individuals from various backgrounds to participate in its ecosystem - which in turn ensures that the DeepTruth system has as little ideological or cultural bias as possible. We are not aware of any other project that utilizes the power of the crowd and artificial intelligence to fight false information in such a manner. The blockchain powered utility token that will enable the decentralized exchange of validated information in the DeepTruth system.

The Rand Corporation, an American nonprofit global policy think tank, has termed the absence of a shared language with which we can discuss our problems as Truth Decay. Truth decay is being brought on by four related trends:

1. increasing disagreement about facts and analytical interpretations of facts and data,

2. a blurring of the line between opinion and fact,

3. the increasing relative volume, and resulting influence, of opinion and personal experience over fact, and

4. declining trust in formerly respected sources of factual information.

These trends have been in existence for some time, but now they have reached critical points. Predictably, the absence of a shared language for political and social discourse has disastrous consequences. Without foundation for discourse, without shared truths, we cannot achieve compromises, and without compromises, our social fabric fractures.

By creating an immutable historical record of all of the validated to be false information, detailing where it originated from, who made changes to it, where it was shared, the DeepTruth DApp will create a “truth provenance” in its blockchain layer which will not only help to refine DeepTruth AIs but will also create a valuable competitive advantage. Amongst other things, those individuals and businesses that are concerned about the impact of false information, like deep fakes, will be able to utilize our services for their timely detection and reporting.   

Current - Nil

One Year - Approx 50,000

5 Years - Over 2 million

As deep fakes become incredibly realistic and easy to produce, their impact will be qualitatively different from what we have seen before. In the near future, we see the DeepTruth system forming a layer of authenticity that will become just as necessary for browsing the web as anti-viruses or ad-blockers are today.

As a result, our key drivers include enabling millions of users to help uphold the truth in a digital age.

To compare another blockchain based browser that focuses on advertising, Brave, who emerged May 31 2017, secured their 1 millionth web browser user in just over 18 months to January 2019. Four months later to April 2019, and they had 5.5million monthly users, an increase of 4.4million, or a growth of 1.1million users per month.

We believe that we can do the same with DeepTruth.

Technical

We have dedicated developers on our project as part of SingularityNET, but will need to scale and grow as the adoption of the DeepTruth Browser increases. We are a worldwide team who work remotely, so there are no issues in welcoming new team members into the fold.

Cultural

Culturally, we are a worldwide team spread from USA, UK, Netherlands, France, India, Pakistan, Australia, and Russia. We all relate to the ideals that DeepTruth offers, and do not feel that there are any cultural issues that would affect DeepTruth.

Market

The market is open to users and validators securing the truth, especially as we will be rewarding users for their identification of deepfakes with the DT Token. Even without a reward, the social karma of being part of the platform will drive users interests.

This has been explained in the section above

N/A

We have a team of 10 volunteer personnel at the moment that work directly on DeepTruth.

We are uniquely positioned to deliver the solution that we are proposing as we are marrying blockchain, AI, and human intelligence together in order to eradicate a problem.

Our team also come from previous managment roles at LinkedIn, operations management of Australian Stock Exchange Listed companies, are top writers on artificial intelligence on Medium, and have established developers that are very active in the global AI and ML communities, attending and sponsoring many hackathons across the world.

We aren't at the moment, but looking to network with the right fit in conjunction with professional guidance from MIT

We will offer Truth as a Service (TAAS) to the following:

Technology platforms such as Facebook will look for 3rd party validation of deepfake media very soon, the recent Nancy Pelosi deepfake was reported to Facebook by a third party, which took 13 hours, and not almost instant as the DeepTruth Browser will offer. The company that was engaged by Facebook to do so was paid $400,000 in 2018.

Entities with a high brand value who look to prevent potential infrignements of themselves or their establishments will need Deepfakes verified as false in an expedited and professional manner.

There are three main verticals planned in the DeepTruth business model:

1. Ad-brokerage on the DeepTruth Browser, with adverts broadcast as trustworthy, whilst broadcast on a trustworthy browser.

2. Software sales on Deep Fake Creation tools.

3. Offering a Deepfake Detection Service to high value entities, corporations, and individuals.

Other options include monetization of AI data sets from user interaction that can be sold to AI companies looking for existing data sets, rather than building themselves.

We are currently determining our forecast costs over the next 5 year with regard to a private seed and series A round funding event.

We would seek funding that aligns with a 20% increase to forecast costs until our revenue model could sustain itself under its own steam and without relying on further funding rounds through equity sale etc.

We believe in alignment of partnerships, and describe in reference to MIT's values below why we are applying to Solve:

Integrity

Our team's focus is not only our own actions, but that of the worlds, hence the creation of DeepTruth. We beleive that history should not be construed by technologies that advocate false occurences, and that a counter technology built on trust is the remedy.

Community

Through the introduction of the Deep Truth Browser, we look to establish a community numbered in the millions, backed by Solves 4 year history and success

Diversity and Inclusion

We believe in diversity and inclusion in every aspect of DeepTruth. We are a globally based team of different races, religions, and sexualities, who have one focus for the human race: the Truth.

Collaboration

Collaraboration wise, DeepTruth's network model is built upon working with people. Without collaboration, which is what MIT is also built on, we are at the mercy of misinformation and deep fakes running rampant. We are open to working with many advisors and partners in order to further bolster our position in securing the truth for our future generations.

Innovation

Through our use of AI, blockchain, and a reward system for users, the pinnacle of success for DeepTruth lies in our continued innovative approach to the development of our technology. Like Solve, we are dedicated in our part in solving global issues, with the emergence of technologically enhanced truth being our primary, collective driver.

To be added and edited later

We would like to apply for the AI Innovations Prize, and we would utilize the prize to further enhance our AI. We would do this by training our AI on broader data sets .............editing required

No

No

No

Yes

No

Perhaps