Stay up to date with the HUMAN project through our bi-monthly newsletter.
A CAPTCHA is a question. What appears an instruction – please click each image containing a car – could equally be phrased as: which images contain a car?
The CAPTCHA serves two purposes: one, to verify the humanity of the user; two, to label data of value to ML practitioners and data scientists.
hCaptcha, the first job Requester on HUMAN, was designed to compensate websites that host the service for the work their users do when they click images correctly. ML enterprises pay; websites who host hCaptcha earn. In creating the service, the team realized that a larger problem needed to be solved: the problem of employing software to automate the management, evaluation, and compensation of work. This was the genesis of HUMAN Protocol; and it led to the evolution of CAPTCHA style questions, both as an on-chain human-verification solution Proof of HUMANity, and as a broader platform for the on-chain launching, management, and response of any question.
If you can ask hundreds of millions to label a traffic light, a crosswalk, or a car, what else could you ask them?
HUMAN Protocol was established to democratize access to these global respondents. The first applications on HUMAN are data labeling applications, such as hCaptcha, Intel CVAT, and INCEpTION, to provide ML with the data required to produce brilliant AI products for the future.
Just as hCaptcha was an example, so are all these applications; while HUMAN Protocol created the infrastructure for these data labeling applications to tackle the problems in AI, in doing so, it tackled the broader problem of employing software to organize and compensate contribution.
The resulting possibilities are limitless.
CAPTCHAs remain a test because, in a test, there is a “pass” threshold; the potential for both correct and incorrect answers. But the potential of a global Q&A has broader application in the objective verification of data, and the subjective collation of opinion.
Opinion has value. More precisely, vast consensus has great value. Any project could benefit from asking hundreds of millions a single question; a question that could provide them actionable business information based, backed up, and verified by data analytics.
Below are three potential use cases. These would either be through CAPTCHA style applications, which expand beyond the ‘click the images containing’ format, and include the ability to type answers, select preferences, options, and answers of all kinds. This has already been demonstrated by the hCaptcha API, which enables this current question/answer response before verifying HUMANity:
Alternatively, projects can loop in their own applications and APIs.
Importantly, the service can specify the profile of the respondent. While a global, random sample could be appropriate, it could be equally appropriate to access users on, for instance, a Spanish-native site to establish what company tagline resonates with that language, background, and culture.
A/B testing has long been the foundation of successful business. Whether a project wishes to compare two, or ten, brand logos, or the one-liner which most resonates with them, they can.
This technology has significant implications for business analytics. Think of a startup entering the chocolate industry. They plan to launch in France, Canada, and Switzerland, and decide to ask respondents from those locations what flavors they most like; or which product they associate with luxury, or whether they eat chocolate at all.
The questions are limited only by the questioner’s imagination. Collecting responses to these kinds of questions would lead to valuable business information.
Questions could involve images. They could be multiple choice, or allow respondents to provide their own answers, as with the text-based CAPTCHAs.
While A/B testing and Q&As are kinds of polling, the ability to collect real-time data while events unfold gives HUMAN Protocol technology a unique advantage. Traditional polling techniques take time. Speed of response opens up new business opportunities, particularly where a first-mover advantage is available, for instance, in prediction markets.
HUMAN has created an application to feed prediction markets with valuable opinion polls to help expedite the settlement of the markets: IMOO. This protects them from arbitration; in an industry where incremental advantages can have sizable impacts, we expect IMOO to be a popular application, defining a new standard for anyone looking to enter the prediction market space.
IMOO is another example of the fundamental proposition of global Q&A capabilities made possible through the Protocol, and made tangible by application. The HUMAN team will continue to work with many partners in the open source community, and beyond, to improve the functionalities and applications of the Protocol.
By enabling any project to access such a global pool of respondents, we hope that HUMAN Protocol can contribute to the growth and excellence of many projects and business solutions.
For the latest updates on HUMAN Protocol, follow us on Twitter or join our Discord. Alternatively, to enquire about integrations, usage, or to learn more about how HUMAN Protocol supports machine-learning technologies, get in contact with the HUMAN team.
The HUMAN Protocol Foundation makes no representation, warranty, or undertaking, express or implied, as to the accuracy, reliability, completeness, or reasonableness of the information contained here. Any assumptions, opinions, and estimations expressed constitute the HUMAN Protocol Foundation’s judgment as of the time of publishing and are subject to change without notice. Any projection contained within the information presented here is based on a number of assumptions, and there can be no guarantee that any projected outcomes will be achieved.