Microsolve Business IT Insights

Balancing AI Staff Policy: Encouraging Use When Security Matters

Written by Dale Jenkins | 5 August 2024 12:01:28 PM

Like many tech-based SMEs, Microsolve has embraced Artificial Intelligence (AI) as an integral part of our workplace. Our internal AI usage policy encourages experimentation and tool usage, provided the supplied data doesn't contain commercially sensitive information or anything that could identify a client or their environment. Pretty standard stuff, right?

We also pride ourselves on an inclusive workplace culture that actively challenges assumptions and industry norms. We love playing devil's advocate, if for no other reason than to stay ahead of our clients' future questions.

Let's explore the challenges that we, along with many other businesses, face in balancing AI adoption, privacy, security, and the needs of an inclusive workforce in the Australian marketplace.

The Nature of AI Tools

AI tools come in various forms, each serving different purposes. Remember, they're only as good as the data sets they have access to for training. These tools excel at parsing, processing, and using data from documents, emails, images, text, chat messages, voice recordings, and web or intranet sites to answer your questions and generate responses. Think of them as your personal, in-the-pocket assistant for countless tasks, saving you precious time.

Encouraging the Use of AI Tools

Encouraging everyone in your organization to harness the power of AI (within policy guidelines) is almost a no-brainer. AI-driven analytics can provide insights from vast datasets, speeding up analysis and decision-making. Pair this with Natural Language Processing, and you've got a tool that can both crunch enormous data sets and communicate the answers in an easily accessible manner. Brilliant, right?

AI and Assistive Technology

Assistive technology has long been a game-changer for those with different abilities. Tools like Grammarly, Natural Reader, Otter, and Fireflies enable anyone to process and convert information from one medium to another. For sight or reading-impaired individuals in the IT industry, these tools are essential, opening up enormous opportunities for inclusivity, creativity, engagement, and productivity.

So, what's the catch? All these technologies require access to data. To convert text to voice, the text needs to be processed. But where does this processing occur? Is the information retained?

Data Privacy and Security

One of the primary concerns with any AI tool is data ownership. Once submitted, can that data be retained and used for future training? The potential for accidental release of confidential or private data is real. Imagine an employee with a text-to-speech tool receiving an email from a client containing sensitive commercial data. Without any action from the employee, that content could become available for future training and retrieval by anyone using the tool. Yikes!

Balancing Inclusivity and Information Security

The Australian Government has rightly emphasized the need for clear and consistent data governance to foster public confidence in AI technologies. Companies must implement stringent data protection measures, but we're also bound by the anti-discrimination Act and need to provide an inclusive workplace. How do we navigate this paradox?

Know the Tools - Know the Rules

Finding a balance is possible. From personal experience, it's about establishing (or reinforcing!) security and privacy expectations with the entire team, then understanding and evaluating the tools required to get the job done. Ideally, each tool in use has already been through a security evaluation (it has, right?) and has privacy and data storage reference links on file. If not, better get on that ASAP!

Conflicts will arise when a tool doesn't meet organizational security/privacy requirements but is essential for employee accessibility. This will happen and will require open, clear communication from all parties to find a solution that works for everyone.

Conclusion

This area is far from trivial. If you don't currently have an AI Acceptable Usage policy, I strongly encourage you to engage with an Employment Policy specialist and a Technology specialist. Together, they can provide guidance on your risks, options, appropriate tools, and approaches to ensure ALL of your team can benefit from the advantages AI offers.

It's time to embrace the future while protecting what matters most!