"Data is the new oil" is an expression that most people have heard, but few understand what it really means.
Data is one of the most valuable assets a business can have nowadays. It's the core of the solution to many business problems and an important driver for productivity and profitability alike.
In fact, many businesses consider data acquisition as an investment and spend a lot of money buying or gathering quality data from platforms that make a lot of money selling such data.
The privacy problem
Regulations like the GDPR in Europe put businesses under a lot of pressure to make sure they can still use data in an efficient way in terms of their strategic goals while also being compliant with data protection policies.
And that trend was already happening even before AI was a thing. The surge in AI development has intensified the demand for data many times, creating a very messy scenario in the privacy landscape.
AI brings some positive developments in terms of privacy such as the ability to anonymize or pseudonymize data, making it harder to trace back to individuals while still useful for analysis. On the other hand, it also brings complications as AI systems often require vast datasets to function effectively, but how this data is used, stored, and who has access to it can be opaque, raising privacy concerns.
These and other challenges have spurred the development of data protection protocols involving technologies such as Artificial Intelligence and Blockchain.
Let's take a quick look on some of the main protocols of this type.
Zero-Knowledge Proofs (ZKPs)
ZKPs are a cryptographic method that allows one party to prove that they know something without revealing any information beyond the proof itself.
One example of this use case is Grass, which allows users to contribute idle bandwidth that is used to scrape public web data to train AI models.
Grass Network uses ZKPs to scrape public data only, and not public information, verifying the data's integrity and origin.
Zero-Knowledge Transport Layer Security (zkTLS)
zkTLS merges the ZKPs concept that we just analyzed with the widely-used Transport Layer Security (TLS), delivering a more secure and private method for data transmission.
One interesting application of this protocol is allowing users to prove their income or employment status to lenders, for example, without revealing any sensitive data such as bank statements.
Trusted Execution Environment (TEE)
A TEE is a secured area of a device's processor that's separated from the normal execution area that protects data and applications from unauthorized access.
Fully Homomorphic Encryption (FHE)
As someone with no experience in the subject, I find this protocol to be the most interesting one out of the ones described in this article.
FHE is a technology that allows computation to be performed directly on encrypted data without needing to decrypt it first.
This type of technology greatly benefits on-chain governance, for example, as it allows voting to occur over encrypted data, ensuring votes remain private, which reduces risks of coercion or bribery
Final thoughts
Data becomes more important and valuable by the day, which creates many opportunities, risks and concerns regarding privacy, integrity and security.
Artificial intelligence is a hungry beast that feeds on tons of data, hence playing a major role in this landscape.
AI can be used to facilitate ethics and privacy in data usage but can also be used to compromise those values and therefore, security-focused protocols are a key piece of this puzzle.