The Next Wave of AI Innovation Needs Your Business Data

In an article published on Business Reporter, Vincent Korstanje, CEO of Kigen, spotlights for business leaders that ongoing investments in cybersecurity at the edge with integrated SIM (iSIM) and eSIM-secure elements that are already the bedrock identity management of cellular loT – act as essential trust anchors to unlock business data-driven innovation on Al, and plan long-term success for Al adoption.

Kigen predicts that given the current pace of companies working on improving AI models, developers could run out of data between 2026 and 2032, according to a study released in June 2024 by the research group Epoch AI.

Over the past year, many of the most crucial web sources have been used for training AI. models have restricted the use of their data, according to a study published in July 2024 by the Data Provenance Initiative, an MIT-led research group.

The real breakthrough that will allow humanity to jump to the next S-Curve is data produced at work declared an article featured in July 2024 by Emergence Capital on Fast Company.

To overcome the limitations and hallucinations of Generative Al and Large Language Models (LLMs) that underpin them, businesses need more control over their data and guidance to prepare for how they will play in the future evolution of Al models. This is the focus of cybersecurity firm Kigen’s comprehensive guide for business leaders on equipping their company with cyber smarts for Al.

Speaking on why security is top of the business agenda, Vincent Korstanje said, “In an Al-powered world, security isn’t a feature, it is a necessity.”

Kigen

By 2028, it's estimated that 50 percent of AI workloads could be moved to the edge. Only 4 percent of businesses say that their business data is ready for AI applications. 71 percent of business executives regard the security of their data and IP as a top concern for AI strategy.

An immediate starting point is at the secure Edge Al, i.e. the processing of data locally from your sensors, devices, and products directly rather than in centralized LLMs. Al is tailored to a business’s unique needs with data unique to that business within the context of its industry.

Sensor-driven data is the most potent way to sense, verify, and add to the integrity of the data based on Al inferences. Further, Kigen’s approach, which extends the GSMA loT SAFE standard to secure enterprise credentials, allows each piece of data to be cryptographically signed and sealed, addressing a better fit to the rising need for data provenance and model explainability.

A refreshing change from the hype and concern around AI, Vincent’s article shows how achieving this is within the means of most companies considering digital transformation. Read the article on Business Reporter’s Digital Transformation special edition or access the full report on cyber smarts for AI.

See coverage of Vincent’s article in the premium French business newspaper Le Figaro.