6 Ways that AI Evolves the Channel

Where is AI headed as it relates to the #channel - AchieveUnite 's panel, called "The AI Driven Channel" aims to tell us. In listening to the session, here are a handful of key findings I was interested in:

1) AI was been going on for a while. In many cases, it's been used behind the scenes and associated with Machine Learning models that do things like help systems categorize when pictures are Chihuahuas and when they're Blueberry Muffins (learned this example years ago at a research conference from Chris Robson ). These are the models that companies use to recommend different products and services to you based on your past activity. The recommend content that's relevant for some N current situation, BUT they ignore the current AI model that has been in the news of late, that digests huge quantities of information to answer questions, create presentations, and design art. Both may leverage Deep Learning.

2) For AI to get traction, it's got to be introduced into every day life. It's got to make it into the business workflow of the everyday consumer of the data. Extended into the channel, AI could have an impact on inventory, manufacturing, incentives, on-the-fly content generation to support sales around available inventory, etc.

3) Deal registration has been really slow to evolve. It's been 'stuck' in it's current state for a long time. It's an area that is highly prone for being impacted by AI. What if sales people can put the deal registration portal on an email thread that 'automatically' enters the deal into the system, and if it meets a set of criteria it can be approved without intervention. That'd be an improvement for sure. The decrease in time spent processing non-scalable information, allows vendors to spend more time on the things that move the needle - building relationships. And, at the same time drives 'in-the-moment' content to consumers using highly segmented data.

4) Predicting the next deal and winning opportunities faster are highly likely through advanced AI models. And, most importantly, the ability to identify potential partner churn becomes more accurate and non-manual. But, it does mean that companies have to build these models for themselves, based on what the model should know about YOU.

5) What's the role of trust in this process? AI, at least currently, isn't always accurate. Can you trust the data? Mostly. AI can't replace the human, quality assurance, layer of the business relationship. Someone still has to train, teach, and check the data. As Allison Bergamo said in the chat commentary, "AI is an augmentation tool, not an easy button." It's still a language model and may have inconsistencies in some situations.

6) Finally, and I thought this comment summed the session well, Daniel Nissan referenced 微软 's about to be released Co-pilot. He mentioned, rightly, that it's not THE "Pilot." It's there to help, guide, improve, and support. It's not there to fly the plane.

联想 Mindmatrix StructuredWeb

Theresa Caragol Jessica Baker Harbinder Khera Daniel Nissan Jeff Taylor