Skip to main content

Will data kill the AI star?

(Image credit: Image Credit: Geralt / Pixabay)

AI is the newest rock star on the technology scene. But just because artificial intelligence (AI) is the hottest new thing, that doesn’t mean it can survive the inherent data challenges that come with it. These challenges include data accessibility, selection, timeliness and trust. So just as “video killed the radio star,” data is threatening to kill the AI star. 

According to a recent Infosys survey, 49 percent of organizations reported that they will not be able to deploy AI because of data challenges. Thirty-seven percent of respondents also cited data integrity as a barrier to getting AI projects off the ground. 

In thinking about some of the obstacles to AI success, I’ve noticed some parallels between AI and a literal rock star. Let’s take a look.

If a musician goes on stage and just starts playing a bunch of notes and singing lyrics that don’t align to the notes or to each other, they will be booed off the stage – which translates to no music contract, no money, no stardom. However, if the musician arranges the notes and lyrics into a harmony that aligns together to inspire a reaction, they will earn fans, cheers, a record contract, perhaps even become ... a rock star. 

Keep in mind that the rock star had to go through a process to get the right lyrics, for the right notes, to produce the right results. AI is the same. So let’s see how we can save your AI star and use it to tackle your business challenges.

Backstage: The crew (data governance) makes the show successful

Data integrity and reliability are once again the underpinnings of any successful business initiative, and AI is no exception. Without the right data in the right place at the right time, your AI star is dead.

So how do you keep AI alive? The best place is with a data strategy that incorporates a data governance program. The data strategy is a plan to improve the ways you acquire, store, manage, share and use data.   

A strong data governance program ensures data is managed consistently and correctly, and is provided to the right business process at the right time. Consistently managing the data via governance will ensure that data being used for AI supports AI. In addition, the rules and details of the data will be documented and applied; thus the integrity and reliability challenges of the data for AI are resolved.

Once the right and accurate data is provided to the AI solution, then AI will do what it does best – make an intelligent decision. However, if the data is incorrect, latent or inconsistent, then the AI cannot make the best decision; its decision-making capabilities are constrained by bad data.

Bouncers: (Data) security and privacy are essential

IT security and privacy are also cited as stumbling blocks for AI success. In the Infosys survey, 45 percent of respondents stated IT security and privacy are roadblocks for AI projects. But again, data governance can help organizations overcome these challenges. Organizations must identify what data is sensitive and establish clear definitions that outline, define and identify this data. They must establish policies that clearly state who has access to, and visibility and usability of, sensitive data for the AI process.  

Organizations should also document what happens to unauthorized use of sensitive data and how someone reports data that has slipped through the cracks. Technology should be put in place that can flag, identify and encrypt sensitive information to ensure privacy while still being included in the AI process. 

Sound familiar? It should, because your company should also have data governance policies and procedures for sensitive data being used in your mainstream processes.

In addition to securing sensitive information, there needs to be traceability of sensitive data used by AI. The information security officers and/or data protection officer need to ensure the policies and procedures being established for securing this information are being met while understanding where sensitive data is being used for the AI decision-making process. 

Let the rock concert begin!

Now that everything is in place, AI is ready to take the stage. Imagine if you will: Our star walks out on the stage. The crowd goes wild in anticipation. Our star picks up the guitar, walks to the microphone, and hits that first note. The lights start strobing and flashing, fire begins shooting from the stage, the crowd’s excitement builds to a crescendo, and the star gives a perfect performance. The fans talk about the concert for months – years, even – and the rock star’s popularity grows. (And so do their fortunes.)

It is important to note that the star would not be a real star without their crew or bouncers. These foundational elements ensure a great concert experience for fans while also protecting the star. The star can perform without worrying if all the lighting, instruments, sound systems, etc., are set up correctly. With their security detail, they don’t have to worry that something unpleasant will happen unexpectedly. This all allows the star to do what they do best: perform and give us a show to remember.

Good, clean data paves the way for successful AI. Everything starts and ends with data. Having the correct information in the right place – and managed properly – will empower your organization to glean powerful insights from your AI techniques and tools, which increases the organization’s popularity with customers and subsequently increase your fortunes.

For AI to earn celebrity status, the right notes (data) must be put together with the right lyrics (AI) for the best results. Add in the crew (data governance) and the bouncers (data privacy), and AI will become a rock star for your organization.   

Kim Kaluba, Senior Product Marketing Manager in Data Management at SAS 

Image Credit: Geralt / Pixabay

Kim Kaluba
Kim Kaluba is a Senior Product Marketing Manager in Data Management at SAS. She has 20 years of experience in data management, including sales, marketing and enablement. Kaluba received her business degree in marketing and management from Stetson University.