Tags
Read time
8 min read
Last updated
April 11, 2025
🔁 Migrate from OneTrust seamlessly with Ketch Switch. Exclusive offer through 4/30/2025.
Ketch Data and AI Governance capabilities will help businesses unlock the value of AI initiatives without compromising privacy and compliance requirements.
Ketch makes consent banner set-up a breeze with drag-and-drop tools that match your brand perfectly. Let us show you.
Generative artificial intelligence (GenAI) is going gangbusters. Since ChatGPT broke into the mainstream in late 2022, GenAI has rapidly evolved from a curiosity into a business game-changer. Companies of all kinds are investing accordingly:
From machine-generated poetry to revolutionary innovations in semiconductor design, GenAI has the potential to disrupt, streamline, and augment virtually every business function across countless different industries.
Part of GenAI’s power lies in its inherent ease of adoption: with prompt-based, natural-language interfaces, GenAI is becoming accessible to millions of new users who were previously shut out of first-gen AI tools.
GenAI has the potential to automate away the busywork that currently consumes the attention of human employees. McKinsey estimates that GenAI tools could ultimately free up 60% to 70% of employees’ time.
The combination of a remarkably low barrier to entry and incredible potential to drive value across countless domains make GenAI a powerful force.
However: there’s another side to the story. Even as companies are investing time, money, and resources in GenAI, three-quarters of global businesses — including tech giants such as Samsung and Apple — have banned or are considering banning their employees from using GenAI tools.
What’s leading so many companies to floor the gas pedal while simultaneously slamming on the brakes? GenAI radically changes companies’ relationship with both consumer and corporate data.
GenAI’s power comes at a price: these transformative technologies have an insatiable appetite for data.
Large language models (LLMs) and other GenAI tools work by detecting statistical patterns in vast pools of training data: OpenAI’s GPT-4 model, for instance, was originally trained using billions of words of text and hundreds of millions of images distilled from a 45-terabyte data dump of the entire internet. Huge amounts of public data — some of it subject to privacy or copyright laws — is baked into the foundation models that underpin virtually all GenAI tools.
Typically, GenAI models need to be re-trained or fine-tuned to ensure they produce appropriate outputs for a given use-case, then further refined and improved once they’re deployed. That requires companies to feed large amounts of additional consumer and corporate data into their AI models.
Acquiring and activating data is already a critical business imperative. But AI makes it even more important: companies will need vast quantities of data to train, fine-tune, and continually refine AI models. If companies lack access to high-quality data, their AI projects are dead on arrival.
Companies already struggle to activate their most valuable data: at present, according to IDC, 44% of business data isn’t effectively captured, and 43% of captured data goes unused. The result: less than a third of the data flowing through today’s organizations is actively utilized to drive value for businesses and their customers.
GenAI tools will change this. As AI technologies grow more commonplace, businesses will increasingly find that all their data — both consumer and corporate — will touch an AI model at some point in its lifecycle. Already, 40% of organizations have thousands of AI models deployed across their operations; that number will only increase in coming months and years.
Once a GenAI tool is deployed, it continues to absorb data as it’s leveraged by end-users. It’s incredibly easy, for instance, for employees to copy-paste sensitive data into AI prompts; among organizations that have faced AI security or privacy incidents, 60% report that their data was compromised by the actions of their own employees.
As GenAI tools grow more commonplace, employees will use them whether their employers like it or not. In fact, Gartner estimates, by 2026 at least 5% of workers covered by GenAI bans will ignore the prohibitions and use the tools to simplify their daily workflows.
The reality is that GenAI bans won’t deliver real data security — but they will stifle innovation and make it harder for companies to keep up with the competition. We need a smarter approach that makes it possible to maintain data privacy and security while still using GenAI to boost productivity and drive business growth.
Combine these challenges with the rapid evolution of both consumer expectations and the global regulatory landscape, and the complexity of the risks facing today’s businesses becomes clear. GenAI makes data more important than ever — but it also makes managing data responsibly, and securing continuing access to data, more difficult than ever before.
As businesses, we need to reconcile these two opposing faces of GenAI:
This requires true privacy-safe infrastructure — giving both consumers and companies the agency and data security they deserve.
By 2028, it’s estimated that over half of companies now building custom AI models will have shelved the projects due to spiraling costs and ever-increasing complexity. Even today, just 54% of enterprise AI projects make it from pilot to production. Data security and privacy are key factors in determining which organizations succeed and which struggle.
At Ketch, we believe that this calls for a new paradigm: privacy-safe AI, grounded in automated technologies that ensure end-to-end data data privacy across the entire AI value chain.
Ketch Enterprise Privacy Agent is a cutting-edge solution designed to align artificial intelligence (AI) initiatives with responsible data governance. It empowers organizations to harness the capabilities of AI while ensuring compliance with data privacy regulations and ethical standards.
By integrating privacy considerations into the core of AI operations, the Enterprise Privacy Agent facilitates the development and deployment of AI models that respect user consent and data protection laws.
This agent operates by embedding privacy controls directly into AI workflows, enabling automated enforcement of data usage policies across various systems. It provides a framework for managing data subject rights, consent management, and data mapping, ensuring that AI models are trained and function within the bounds of established privacy preferences.
Through its comprehensive approach, the Ketch Enterprise Privacy Agent not only mitigates legal and reputational risks but also fosters trust among users by upholding their data rights in AI-driven processes.
By 2026, organizations that successfully build out privacy-safe AI solutions grounded in transparency, trust, and data security will realize a 50% gain across key metrics including user acceptance, technology adoption, and total business value. Companies that fall short on those measures, of course, will be left in the dust.
At Ketch, we know the AI arms race isn’t a sprint — it’s a marathon. The goal isn’t simply to surge ahead at any cost. It’s to build a stable foundation that ensures permissioned and privacy-safe data flows through all your AI operations, and that your AI projects aren’t merely a flash in the pan.
The bottom line: privacy-safe data is the key to unlocking lasting value for your customers, and lasting growth for your business. In a world that’s being redefined by GenAI, putting privacy and security at the heart of your AI projects is the key to building competitive advantage and finding enduring success.
Learn more about our privacy-safe AI solutions → Request early access.
