Data fabric supports efficiency, privacy in AI era

Johannesburg, 02 Dec 2021
Read time 2min 20sec
Comments (0)
Tatenda Gatawa, Data and AI Sales Leader, IBM Southern Africa.
Tatenda Gatawa, Data and AI Sales Leader, IBM Southern Africa.

Increased AI use has the potential to raise privacy issues or stifle data flow with restrictive policies meant to prevent data misuse. 

So says Tatenda Gatawa, Data and AI Sales Leader at IBM Southern Africa. “Thankfully, data privacy and data use do not have to be mutually exclusive, he says. “By introducing a data fabric, data privacy and lineage tools, and consistent reporting and dashboards, organisations can support both efficiency and compliance.”

IBM explains that a data fabric is a data management architecture that can optimise access to distributed data and intelligently curate and orchestrate it for self-service delivery to data consumers. It is agnostic to data environments, data processes, data use and geography, while integrating core data management capabilities. With a data fabric, business users and data scientists can access trusted data faster for their applications, analytics, AI and machine learning models, and business process automation.

“Data fabric solutions enable self-service access to all data from any location, without requiring it to be moved or transformed,” Gatawa says. “A data fabric decreases the danger of data corruption and loss from transportation by combining data virtualisation and governance, as well as providing a shared metadata catalogue.”

Says Gatawa: “This presents a great opportunity for organisations and whole industries alike. Gartner predicts that by 2024, “the use of data protection techniques will increase industry collaborations on AI projects by 70%”.

He notes that user concerns about incorrect data access can be resolved with the right data fabric, which can help boost compliance without incurring needless delays, thanks to access controls and dynamic masking of sensitive data. Similarly, the provenance of data and models can be tracked automatically so that the business can better determine whether data and models are being used correctly. Users gain even more information at the same time, which aids in their decision-making about which datasets to use.

“Keeping track of regulatory compliance and model performance is simplified by the data fabric. Organisations require tools that track complex regulatory information from multiple sources, deduplicate it and apply it automatically. UIs should also assist users in making the right regulatory decisions in the moment, without extensive training. And, as always, compliance must be tracked via dashboards and audits to identify deviations before they become a problem. AI models should be tracked and evaluated not only for performance, but also for explainability and bias,” Gatawa says.

Login with
4 hours ago
Be the first to comment
See also