THE PROBLEMS FACING THE DATA ECONOMY
There are four major problems facing the data economy:
- A widening AI data gap between the haves and have-nots
- Ubiquitous fake and unauthentic data destroying the usefulness of any algorithm
- Unsustainable data feeds breaking production systems when they go offline
- Unfair revenue distribution for the original data producers
AI Data Gap between Haves and Have-nots AI requires vast volumes of data that not every company has access to. The data exists but is often scattered throughout various industries and made up of different types. This makes it difficult for small companies to procure the volumes of diverse data that they need. Large companies are far more immune to this problem because they have the resources to collect and store the volumes of data necessary for their AI activities. Google, for example, has access to vast amounts of data via its own products: conversation and purchase history from Gmail, location history from Google Maps and mobile activity from the Android OS. It is able to take a much more
holistic view as a result of having so many users. Companies like this are the ones producing AI innovation at a much larger scale, solving bigger problems and producing better products.
Where the problem lies is in the ability for everyone else to gain access to the data that they need to compete. It is either prohibitively expensive or not available for purchase in the volumes required. In some cases, the data exists in small fragments that are contained by entities that do not know what to do with them.
The reality, however, is that there is a ton of data out there. Devices, sensors, people, businesses and any other technology that one can think of are producing vast amounts of data every second. However, unless one has the resources to harness it, the data remains out of reach.
What is missing is a solution to organise and make this data available to everyone. Everyone should be able to achieve the best results from his/ her algorithms and, therefore, produce better solutions for the world.
Where the problem lies is in the ability for everyone else to gain access to the data that they need to compete. It is either prohibitively expensive or not available for purchase in the volumes required. In some cases, the data exists in small fragments that are contained by entities that do not know what to do with them.
The reality, however, is that there is a ton of data out there. Devices, sensors, people, businesses and any other technology that one can think of are producing vast amounts of data every second. However, unless one has the resources to harness it, the data remains out of reach.
What is missing is a solution to organise and make this data available to everyone. Everyone should be able to achieve the best results from his/ her algorithms and, therefore, produce better solutions for the world.
Fake and Unauthentic Data It is a universal truth that where there is money to be made, people will try to game the system. This is no different in the data economy. Data is easy to fake, copy and misrepresent. This makes it difficult for Data Consumers to properly vet data when purchasing it from third parties.
When purchasing data, Data Consumers want to follow applicable laws and regulations. They do not want to be unethical in their business practices. There is too much at stake if they should be found engaging in unethical practices. Hence from a business perspective, they do not want to pay for data that is not authentic. From a regulatory perspective, they need to know where their data comes from.
Data Vendors are not always so scrupulous. Some are happy as long as they make their money. Data Consumers, therefore, are not always getting the authentic data that their systems depend on. Unauthentic data input equals a poor resulting output. It is the case for data-driven business decisions, algorithmic trading, AI/ Machine Learning applications and oracles for smart contracts. If the data is false, the consequences can be dire.
Unsustainable Data Ecosystem Free data is not sustainable. No entity can continually produce data over the long-haul if they are not being compensated fairly for it, either directly or indirectly. Regarding data that is exchanged for monetary compensation—without proper revenue sharing, individuals and businesses would not be able to keep their doors open and continue to provide data if they are not compensated fairly. This is critical for keeping data streams diverse and authentic; things like IoT sensors properly maintained. Ultimately, it is the small Data Consumers that suffer the most because they lose access to these data streams and may not have ready access to alternatives.
What is needed is a sustainable ecosystem where producers are incentivised to provide authentic data and buyers are willing to pay for it.
When purchasing data, Data Consumers want to follow applicable laws and regulations. They do not want to be unethical in their business practices. There is too much at stake if they should be found engaging in unethical practices. Hence from a business perspective, they do not want to pay for data that is not authentic. From a regulatory perspective, they need to know where their data comes from.
Data Vendors are not always so scrupulous. Some are happy as long as they make their money. Data Consumers, therefore, are not always getting the authentic data that their systems depend on. Unauthentic data input equals a poor resulting output. It is the case for data-driven business decisions, algorithmic trading, AI/ Machine Learning applications and oracles for smart contracts. If the data is false, the consequences can be dire.
Unsustainable Data Ecosystem Free data is not sustainable. No entity can continually produce data over the long-haul if they are not being compensated fairly for it, either directly or indirectly. Regarding data that is exchanged for monetary compensation—without proper revenue sharing, individuals and businesses would not be able to keep their doors open and continue to provide data if they are not compensated fairly. This is critical for keeping data streams diverse and authentic; things like IoT sensors properly maintained. Ultimately, it is the small Data Consumers that suffer the most because they lose access to these data streams and may not have ready access to alternatives.
What is needed is a sustainable ecosystem where producers are incentivised to provide authentic data and buyers are willing to pay for it.
Fair Revenue Distribution Producers of the original data sources have it the worst with respect to revenue distribution. They need to be incentivised to continue producing data, yet more often than not they are paid just once for the data that they provide. It is the Data Vendors that have the ability to resell the same data again and again. There is no way for the producers to find out what happens to the data downstream, where it goes and for what purpose. What this does is cast an opaque layer over the data, so that the producers have no idea of how much money they are owed.
Stakeholders What each of the above problems does is plague the stakeholders of the data economy: the consumers, the vendors and the entities producing the data.
Data Consumers Organisations are embracing data analytics, data science, machine learning and AI in more sophisticated ways than ever before. They are either using their own in-house capacity or looking to firms specialising in data.
In either case, organisations are likely to go through a data adoption cycle similar to the Gartner Hype Cycle:
Stakeholders What each of the above problems does is plague the stakeholders of the data economy: the consumers, the vendors and the entities producing the data.
Data Consumers Organisations are embracing data analytics, data science, machine learning and AI in more sophisticated ways than ever before. They are either using their own in-house capacity or looking to firms specialising in data.
In either case, organisations are likely to go through a data adoption cycle similar to the Gartner Hype Cycle:
- Data Trigger – “I need data! Where can I get it?”
- Peak of Inflated Expectations – “I will pay you anything for it!”
- Trough of Disillusionment – “Wait, this data doesn’t really do what I want!”
- Slope of Enlightenment – “If we improved the data’s x, y, z, it will work wonders!”
- Plateau of Productivity – “Okay, I now have the data from multiple sources at a price I can afford.”
Depending on the current state of an individual market segment, companies might find themselves in multiple stages at once. It is not always a perfectly linear progression.
For example, in the Mobile Location Intelligence data segment, it can be seen that most participants are currently in stage 3 or 4. Conversely, in the broader data economy, many companies are only starting to learn about the challenges that exist with their current data sources and find themselves transitioning from stage 2 to stage 3.
Stage 3 is the most critical, as it can make or break a company and its solution. If they are unable to obtain authentic data, they will be forever chasing the dream of using data to solve a real problem.
Once companies reach stage 5, provenance becomes the highest priority. They need to be sure that they are paying for authentic data that will support the business
For example, in the Mobile Location Intelligence data segment, it can be seen that most participants are currently in stage 3 or 4. Conversely, in the broader data economy, many companies are only starting to learn about the challenges that exist with their current data sources and find themselves transitioning from stage 2 to stage 3.
Stage 3 is the most critical, as it can make or break a company and its solution. If they are unable to obtain authentic data, they will be forever chasing the dream of using data to solve a real problem.
Once companies reach stage 5, provenance becomes the highest priority. They need to be sure that they are paying for authentic data that will support the business
For more information, please visit links below
Website :https://www.quadrantprotocol.com/
ANN Bitcointalk: https://bitcointalk.org/index.php?topic=3676988.0
Whitepaper: https://www.quadrantprotocol.com/whitepaper.pdf
Facebook: https://www.facebook.com/quadrantprotocol/
Twitter: https://twitter.com/explorequadrant
Telegram: https://t.me/quadrantprotocol
Website :https://www.quadrantprotocol.com/
ANN Bitcointalk: https://bitcointalk.org/index.php?topic=3676988.0
Whitepaper: https://www.quadrantprotocol.com/whitepaper.pdf
Facebook: https://www.facebook.com/quadrantprotocol/
Twitter: https://twitter.com/explorequadrant
Telegram: https://t.me/quadrantprotocol
AUTHOR
Bitcointalk Username: anja96
Bitcointalk URL: https://bitcointalk.org/index.php?action=profile;u=1503069
Bitcointalk Username: anja96
Bitcointalk URL: https://bitcointalk.org/index.php?action=profile;u=1503069
Comments
Post a Comment