Unified communication platforms are going all in on AI, integrating artificial intelligence and machine learning to enhance user experiences. And generative AI—which ‘learns’ from historic data to generate new data, including text, images and video—is taking it up a notch.
When executed properly, generative AI can accelerate productivity. But it can also create challenges around security, compliance and ethics.
Going back a few years, the shift to the cloud was a game-changer for unified communications (UC). For users, it brought all communication channels onto a single platform, saving time and increasing productivity. It also streamlined operations by integrating with business apps, such as customer relationship management (CRM) systems.
For IT teams, it shifted upfront capital costs to a monthly operating expense. No longer was it necessary to deploy and maintain complex (and costly) on-prem equipment. Plus, it could be scaled up and down to adapt to demand, helping to optimize costs.
But the game is changing once again, thanks to AI—and, in particular, generative AI. Artificial intelligence isn’t new to UC. We’re already using it to add a virtual background—or blur the clutter of our home office—during video meetings.
How generative AI is changing the game
But that’s evolving, thanks to large language models (LLMs). Now, UC comes with even more advanced features, such as the ability to generate meeting summaries, conversation prompts and even sentiment analysis.
This can help with ‘meeting fatigue,’ so users can read a summary or synopsis of a meeting they attended (or one they didn’t), or even a long chat thread. The AI can also summarize key points, such as any important decisions that were made during the meeting or any follow-up actions required.
With generative AI, users can also provide a prompt—a question or statement—to generate a relevant response, such as composing chat responses or coming up with meeting agendas. It’s also improving chatbots, so they’ll be able to ‘learn’ and evolve, providing more helpful answers, and even use natural language to communicate.
Generative AI can even be used to leverage the capabilities of a UC platform, such as describing the steps involved in a task—or even executing those steps on your behalf, if it’s integrated with your UC platform.
Here are some key use cases for AI in UC:
- Virtual assistants: AI-powered chatbots and virtual assistants can provide instant responses to customer inquiries, even in multiple languages, creating a more personalized experience.
- Speech analysis: AI can analyze speech patterns and detect customer ‘sentiment’ to improve the overall customer experience.
- Speech-to-text transcription: AI can process speech in real time to produce transcripts from phone calls and meetings.
- Real-time translation: In addition to transcribing meetings audio in real time, AI can also translate the transcripts for a more inclusive and global communication.
- Meeting summaries: Aside from taking notes at meetings, AI can summarize those notes, so participants can easily see key takeaways.
- Predictive maintenance: From an operational standpoint, AI could help to predict operational issues or delays, reducing downtime.
Risks on the road to AI
Generative AI is often compared to having an intern or an assistant. But, just like a human intern or assistant, it’s not perfect, and it can make mistakes. There are many challenges around the use and proper execution of generative AI, so it’s important that employees are trained and educated on both the benefits and risks.
- Security and privacy continue to be major concerns. Guardrails should be put in place to ensure that proprietary or private data isn’t used to ‘train’ generative AI, bringing it into the public sphere—which could have compliance and ethical repercussions.
- The use of generative AI could potentially open up your organization to new attack vectors and create new security vulnerabilities. After all, cyber attackers are also using generative AI to become more efficient.
- Another major concern is accuracy. The data that fuels generative AI needs to be ‘clean’ data, meaning it’s high-quality and accurate. If models are trained on poor data, they’ll provide poor results.
- While AI-enhanced chatbots are meant to improve the customer experience, if it’s not executed properly—say, the data is inaccurate—that could make the customer experience worse, not better.
Ultimately, AI in unified communications can boost workplace productivity and make communications more effective. By automating manual tasks, it can also free up workers to engage in more meaningful, value-added work.