Artificial Intelligence: The Real Deal
AI is just another stage of the automation that communications have been undergoing since the beginning of the IP era.
You might already be getting a little tired of the hype around artificial intelligence (AI), and its potential effect on enterprise communications and collaboration. Most of us have a love/hate relationship with hyped technology: On the one hand, you get sick of hearing about it and witnessing people tie themselves into knots trying to associate their companies and technologies with the hot new thing. On the other hand, if there weren't any new things worth being hyped, you'd be worried about the future of the industry. So we can be grateful that technologies that excite people also seem likely to come to bear on the technologies and systems we deal with.
When it comes to AI, though, I think the hype has a negative impact beyond just testing people's patience and B.S. detectors. AI, frankly, is scary, in a way that the Session Initiation Protocol (SIP) never was. Elon Musk never worried publicly that SIP Fuzzing attacks could end human life on Earth. At least in the public perception, there's a mystery about AI that denizens of our industry aren't accustomed to.
But when you break it down, AI is actually likely to be incorporated in enterprise communications/collaboration systems over a long period of time, in stages that seem very intuitive as we move from one to the next. My colleague Michelle Burbick demonstrates what this progression might look like in her recent No Jitter post based on an interview with Jonathan Rosenberg, the technology guru behind the newly introduced AI-powered Cisco Spark Assistant tool. (I'm really happy to note here that Jonathan is going to be Cisco's keynoter at Enterprise Connect Orlando 2018; he's that rare combination of an amazing technical expert who's also an awesome, relatable speaker.)
Jonathan describes the process as five levels of AI, starting with fairly simple tasks (which are of course quite complicated to execute on the back end). It then works up through adoption of natural language processing (NLP), and on to ever-more-proactive steps from the AI, with level five still being a decade out from today. (For No Jitter contributor Zeus Kerravala's bleak but funny line about the AI endgame, check out his post on Spark Assistant.)
When you break it down like this, AI is just another stage of the automation that communications have been undergoing since the beginning of the IP era. SIP may not have been scary, but it did let communications interfaces offer the ability to escalate the same "call" from text to voice to video in one continuous session. That doesn't seem like much now, but it's a feat that Bell Labs would have been proud to claim in its glory days. In the end, it's just a matter of making the tools more useful and usable.
So a chunk of our Enterprise Connect program will touch on AI in one way or another. For starters, Brent Kelly of KelCor is leading a deep-dive session devoted to helping you understand AI and how it'll impact the communications world. And most of what we'll talk about in our brand-new Speech Technologies track will relate to AI, NLP, and their sibling technologies. And since AI is already starting to impact contact centers, you'll find relevant discussions in that EC Orlando track as well. If you want to see a list of all our sessions that will at least have an AI element to the discussion, you can go to the Conference Program page and select AI from the Hot Topics list (and while you're at it, register now using the code NOJITTER and you'll save an additional $200 off the Advance Rate or get a free Expo Plus pass.)
So I for one welcome our new opportunity to talk about AI and see it in action. I embrace the future. What could go wrong?
- Check Out the Enterprise Connect 2018 Program!
- Is Slack Bound for Your Enterprise?
- Charting Cisco's AI Roadmap