As businesses compete to implement AI, many people find that the success of their projects directly depends on the quality of their data. This dependency stalls many ambitious initiatives and does not create it beyond the experimental proof-of-concept stage.
So, what’s the secret behind turning these experiments into real revenue generators? AI News has caught up with Martin Frederik, the Netherlands, Belgium and Luxembourg regional leader of Data Cloud Giant Snowflake.
“There’s no AI strategy without a data strategy,” Frederik says. “AI apps, agents, and models are as effective as the data being built, and even the most sophisticated models can be scarce without a unified, well-governed data infrastructure.”
Improving data quality is key to success in AI projects
This is a familiar story to many organizations. Promising proof of concepts will impress your team, but it will not translate them into a tool to make money for the company. According to Frederick, this often happens as leaders treat technology as their ultimate goal.
“AI is not a destination, it’s a way to achieve business goals,” advises Frederik.
When a project gets stuck, it usually becomes some common perpetrators. The project really doesn’t match what the business needs, the teams aren’t talking to each other, or the data isn’t confusing. It’s easy to be discouraged by statistics that suggest that 80% of AI projects will not reach production, but Frederick offers a different perspective. This isn’t necessarily a failure, he suggests, but it’s “part of the maturation process.”
For those who make the foundation right, the reward is very realistic. A recent Snowflake survey found that 92% of companies are already benefiting from AI investments. In fact, with every pound they spend, they are reclaiming £1.41 with cost cuts and new revenue. The key Frederik Repeats is that from the start, it has a “secured, governed, centralized platform” for data.
It’s not just about technology, it’s about people
Even with the best technology, AI strategies can be flat if the company’s culture is not ready. One of the biggest challenges is to introduce data to the hands of not only the small number of data scientists selected, but also to everyone who needs it. To make AI work at scale, we need to build a strong foundation on “people, processes, and technology.”
This means breaking down the barriers between departments and creating high-quality data and AI tools that everyone can access.
“With proper governance, AI becomes a shared resource rather than a siloed tool,” explains Frederik. When everyone works from a single source of truth, the team stops discussing whose numbers are correct and starts making faster and smarter decisions together.
Next leap: The reason for itself is AI
The true breakthrough we see now is the emergence of AI agents that can understand and infer all kinds of data at once, regardless of the quality of the structure. From neat rows and columns in a spreadsheet to unstructured information about documents, videos and emails. This is a huge step forward considering this unstructured data accounts for 80-90% of typical company data.
The new tool will allow staff to simply ask complex questions in English and get answers directly from the data, regardless of their technical skill level.
Frederick explains this as a move towards what he calls “goal-oriented autonomy.” Up until now, AI has been a helpful assistant you have constantly had to direct. “You ask the question, you get the answer. You ask for the code, you get the snippet,” he points out.
Next-generation AI is different. It can give agents complex goals and know the steps needed on its own, from writing code to drawing info from other apps to providing a complete answer. This automates the most time-consuming parts of the data scientist’s job, such as “boring data cleaning” and “repetitive model tuning.”
result? It releases your brightest mind and focuses on what’s really important. This will help people “from practitioners to strategists” and increase the real value of the business. That’s just a good thing.
Snowflake is an important sponsor of the year AI & Big Data Expo Europe There is also a range of speakers that share deep insights during the event. The Swing by Snowflake booth is Stand Number 50, and we heard more about making Enterprise AI simple, efficient and reliable.
reference: Public Trust deficits are a major hurdle for AI growth
Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo in Amsterdam, California and London. The comprehensive event is part of TechEx and will be held in collaboration with other major technology events. Click here for more information.
AI News is equipped with TechForge Media. Check out upcoming Enterprise Technology events and webinars here.
