Can this Nvidia-Backed Start-Up Change AI Training Forever?

Data Guardians Network has secured a place in Nvidia's selective Inception programme, marking a significant development for the four-month-old start-up focused on AI data governance.

The company, which positions itself as building the world's first Provable Data Governance Framework, joins a programme designed to nurture start-ups that are advancing AI and data science technologies.

A decentralised approach to AI training data

The work that Data Guardians Network does all centres on sourcing AI training datasets from decentralised human contributors, rather than traditional web scraping methods.

The platform operates through gamified interfaces across mobile, web and Telegram applications, where users can contribute data and receive immediate payments in USDT cryptocurrency.

The company covers multiple data types including image recognition, voice and video understanding, and lip-sync data, targeting enterprise clients requiring verified datasets.

"Joining Nvidia Inception is a major milestone in our mission to build the trusted data layer for AI," says Johanna Cabildo, CEO of Data Guardians Network.

Market positioning and enterprise validation

Although Data Guardians Network is less than half a year old, it has raced out of the blocks.

The start-up has already secured US$5m in pre-seed funding and has recorded some huge business successes, like its partnership with DroppGroup.

This particular collaboration saw D-GN train AI models that are aware of the risks of intellectual property infringement, using five million data points covering patents, trademarks and licensing disputes across more than 60 jurisdictions.

Then, DroppGroup deployed these models with some of its massive enterprise clients like Saudi Aramco and Cisco, using platforms such as LLaMA and aMiGO for real-world production environments.

The team at DroppGroup ended up being thrilled with the results.

"D-GN isn't just a crowdsourcing tool – it's live data infrastructure,” says Areeb Masood, Head of Enterprise Deployment at DroppGroup.

“It let us continuously train models like LLaMA-3, while live deployments with our clients stayed accurate, audit-proof and protected in real time."

The case study demonstrates Data Guardians Network's ability to provide real-time, auditable data streams for regulatory compliance in high-stakes sectors including energy, technology and defence.

By regarding human experts as a crucial part of AI’s future, D-GN not only gives its data a higher degree of reliability, it helps assuage concerns about the dwindling role humans might play in an era of more extensive automation.

"We're creating the job role of the future, establishing human-verified data to make AI systems trustworthy and useful," Johanna explains.

What next for Data Guardians Network?

The partnership comes as enterprises face increasing scrutiny over AI training data sources and ethical considerations around dataset compilation.

Traditional AI training has relied heavily on scraped internet content, raising questions about consent, bias and data quality that Data Guardians Network's approach seeks to address.

The company's emphasis on diverse, human-verified datasets reflects growing industry awareness of the need for more accountable AI development practices.

Nvidia's Inception programme typically selects start-ups with the potential to make a significant impact in the market. Naturally, this will fill the team at D-GN with a great deal of confidence going forward.

The firm’s inclusion in Nvidia’s programme will give the team access to a vast amount of technical resources and market networks, which could prove decisive in its expansion over the coming years.

Still, the leadership team at D-GN emphasises the importance of the journey.

“It’s been a whirlwind 4 months for D-GN,” says Richard Johnson, the firm’s COO. “From zero, to our US$5m raise, to Nvidia Inception.”

“This isn’t just about speed though. It’s proof that building the trusted data layer for AI is no longer optional, it’s urgent. 

“AI won’t cross from mimicry into reasoning without provable, human-verified data infrastructure.”