Earlier in October, I defined Cognitive Technologies and provided an introduction to the different approaches of Cognitive, a market growing by 65% CAGR according to IDC. In June 2016, the WhiteHouse published a formal request for information, or RFI, about the possibilities and risks of AI.

IBM ran with this consideration and published Preparing for the Future of AI . Their response answers the formal RFI considerations in the White House RFI and recommends policy makers engage to: (1) Facilitate a fact-based dialogue on the capabilities and limitations of AI technologies, (2)Develop progressive social and economic policies to deploy AI systems for broad public good, (3) Develop progressive education and workforce programs for future generations and (4) Invest in a long-range interdisciplinary research program for advancing the science and design of AI systems.



The new IBM-MIT Laboratory for Brain-inspired Multimedia Machine Comprehension is a multi-year collaboration looking at the problem of computer vision and audition, led by Jim DiCarlo, head of MIT’s Department for Brain & Cognitive Science. Using virtual neural networks modeled on how our own real-life neural networks operate, researchers have produced all kinds of interesting advances in how computers interpret the world around them.

The MIT partnership is one of several IBM has established lately; the company’s VP of Cognitive Computing, Guru Banavar, is also pursuing AI in decision making, cyber security, deep learning for language, etc. All together the group of partnerships comprises what’s called the Cognitive Horizons Network. More information can be found here.


Generating speech from a piece of text is an important task undertaken by computers, but it’s pretty rare that the result can be mistaken for ordinary speech. Now with WaveNet, a new technique from researchers at Alphabet’s DeepMind takes a completely different approach, producing speech and even music that sounds eerily like the real thing. Early systems used a large library of the parts of speech (phonemes and morphemes) and a large rule set that described all the ways letters combined to produce those sounds.

The pieces were joined, or concatenated; creating functional speech synthesis that can handle most words, albeit with unconvincing cadence and tone. Later systems parameterized the generation of sound, making a library of speech fragments unnecessary. WaveNet, as the system is called, takes things deeper. It simulates the sound of speech at as low a level as possible: one sample at a time. That means building the waveform from scratch — 16,000 samples per second. To see the detailed research paper, press here on WaveNet.


At Dreamforce 2016, SalesForce introduced Data Science for everyone with the introduction of Einstein. Salesforce Einstein is artificial intelligence (AI) built into the core of the Salesforce Platform, where it powers the world’s smartest CRM. It delivers advanced AI capabilities to sales, service, and marketing — and enables anyone to use clicks or code to build AI-powered apps that get smarter with every interaction. Now, everyone in every role and industry can use AI to be their best.

The addition of analytics and machine learning has become a key strategy to Salesforce as it bids to build on its cloud offerings. California based PredictionIO, a recent acquisition, is getting immediate access to the entire Salesforce clouds. PredictionIO now has 8,000 developers creating over 400 apps. PredictionIO’s open source technology will continue to be free to all users. To mark the Salesforce deal it is to dropping the PredictionIO Cluster software fee on AWS Cloudformation, which will is now free for the first time in the company’s history.

Helpful Definitions on AI are also here.


SalesChoice is a Certified ISV Partner of SalesForce and our company has successfully integrated with SalesForce EINSTEIN’s WAVE Business Intelligence products. We are making a commitment to continue to build our solutions with EINSTEIN offerings. Our vision is that eventually there will be a global AI Highway connecting the learning signals and patterns across all AI interaction environments, much like the Telecom industry laid fiber down everywhere to create our voice connections and subsequently our internet connections, we are now entering the AI WAR for AI neural network connectivity everywhere.

SalesChoice AI approaches in our deployments are slightly different than SalesForce’s as we bring our Chief Data Scientists to every client engagement to fine-tune our client’s models so they have a UNIQUE AI model relevant to their business.

SalesForce believes that a Data Scientist won’t be needed in most cases. We are not sure about this point yet.

In time, this may be feasible, but there are so many unique nuances to a company’s business that this assumption will require company’s to tread with caution to ensure there is validity in the insights being derived. A company may not need a Data Scientist on staff full time, but to assume the right patterns are relevant and all statistically valid will require checks and balances.

Our experiences have been that AI methods still need more human touches along the AI Bread Crumb journey to make accurate predictions with valuable insights. We have just finished a major global test with one of the largest SalesForce clients in the world and found even though there were hundreds of patterns being logged in SalesForce to predict sales outcomes accurately there needs to always be some AI engine fine tuning to ensure it is meeting the business needs.

There remains a lot of bad data in CRM systems still today, despite the thirty years of hammering on sales professionals to put in better data, or use data cleansing approaches.

We are now installed in major production environments in some of the world’s largest companies using SalesForce and in every company we have found there is customization needed of the software experiences to bring world-class predictive accuracy and insights.

There is a unique set of culture and behavioral and change management dynamics that we need to be acutely attune to in using AI and ML Approaches, and this is an area that SalesChoice is committing to differentiate ourselves with our clients.

What will the Future Hold?

I personally think the future of AI for Sales is going to be in Cognitive Sciences decoding the Talent behaviors and customer interactions inside and outside the CRM, with full signal detection. Email is still a killer APP in business but it also has become one of the biggest productivity wasters of sales professionals as well.

The real questions we need to stop and ask ourselves in Sales AI approaches are:

  • What behaviors do I want my sales professionals to engage in with our customers?
  • Based on (1) how can we monitor a desired baseline against the performance-desired outcomes and improve the gaps?
  • How will our Sales professionals feel if every interaction they do in life is recorded and monitored? Does this motivate the human spirit to advance mankind or is 1984 George Orwell and the Rise of The Machines going too far?

If we are all not careful in AI Cognitive Science Methods – we could find many companies immersed in being driven by signals that are not really that important in securing a customer commitment to do business with.

There must be a balance between AI /ML and human approaches – Science and Art need harmonization, and this is where our market positioning will always be with SalesChoice.

Recently profiled by IDC, Gartner , and having won the recent Silicon Review as one of the TOP 50 in North America, we are increasing our product and market positioning.

How to Contact Us

We welcome a call from you at (647)477-6254 or call my cell directly at (416)230-6538, or cindy@saleschoice.com. You can also book a Data Scientist coaching session on Cognitive Technologies by emailing us at info@saleschoice.com, or book a demo to learn more about our product offerings here.