Sales AI from Uniphone, Sybill, Zoom detects emotions

Hello and welcome to Protocol Enterprise! Today: how companies are selling AI-powered “emotional intelligence” detectors to thirsty salespeople, why Atlassian is still struggling to get hundreds of customers back online, and a potential breakthrough for quantum machine learning.

Twirl up

There’s arguably never been a better time to be a software developer, and while that’s great for the developers themselves, it makes life very difficult for companies that need to hire developers. According to new Mulesoft research from Salesforce93% of companies struggle to retain their good developers over the past two years, and 86% struggle to hire them in the first place.

AI seconds this emotion

Virtual sales meetings have made it harder than ever for salespeople to read the room. So some well-funded tech vendors are stepping in with their own bold sales pitch: that AI can not only help sellers communicate better, but also detect the “emotional state” of a transaction — and the people to whom they sell.

Sales and customer service software companies, including Uniphore and Sybill are building products that use AI to help humans understand and respond to human emotions. Zoom virtual meeting powerhouse also plans to provide similar functionality in the future.

  • “It’s very difficult to build rapport in a relationship in this type of environment,” Tim Harris, director of product marketing at Uniphore, said of the virtual meetings.
  • The company sells software that tries to detect if a potential customer is interested in what a salesperson has to say during a video call, alerting the salesperson in real time during the meeting if someone seems more or less engaged in a particular subject.
  • The system, called Q for sales, can indicate that a potential customer’s sentiment or level of engagement improved when a salesperson mentioned a particular product feature, but then dropped when the price was mentioned.
  • Sybill, a competitor, also uses AI to try to analyze people’s moods while on a call.

But the simple request to record a virtual conversation can alter a customer’s attitude, said Grace Briscoe, senior vice president of customer development at digital advertising company Basis Technologies.

  • “As soon as this recording alert pops up, it puts people on notice,” she said. “I think that would be off-putting for customers; they would be less frank. I don’t think that would be conducive to the type of relationship building that we want to do.”
  • While some sales meeting attendees might be uncomfortable being recorded, others will be more open to it, said Josh Dulberger, chief product, data and AI officer at Zoom.
  • “Part of it is the culture of the sales team,” he said, noting that registration might not be tolerated when selling to more sensitive sectors such as financial services.

Zoom, the king of virtual meetings, said on Wednesday it was introducing new features called Zoom IQ for Sales which provides sales meeting hosts with post-meeting conversation transcripts and sentiment analysis.

  • “We look at things like speaker cadence and other factors in the linguistic approach to try to disentangle one speaker from another,” Dulberger said.
  • Currently, Zoom’s new features for sales reps don’t gauge real-time sentiment in a meeting. Instead, they provide post-meeting analysis.
  • “You’ll be able to gauge that they weren’t very engaged,” he said, noting that sales reps aim for balanced conversations where customers talk as often as a sales rep.

What companies like Uniphore and Sybill are doing goes further as customer service prompts.

  • Uniphore and Sybill aim to monitor human behavior in real-time video calls. And they bet even seasoned salespeople can benefit from the guidance of their AI emotional coaching.
  • Dulberger said Zoom also has active research underway to integrate emotional AI into the company’s products going forward.
  • Briscoe said she recognizes the potential value of emotion-based technologies as management tools to help determine which sellers might be having trouble.
  • However, she said, “companies should hire people who have some level of emotional intelligence. If our team members can’t read that someone got disinterested, those are basic people skills I don’t know why you would need AI [to facilitate].”

Although AI emotional guidance appeals to some sales teams, his validity is questioned.

  • “The claim that a person’s inner state can be accurately assessed by analyzing that person’s face is based on flimsy evidence,” Kate Crawford wrote in a 2021 article in the Atlantic.
  • “We are able to look at faces and classify them into different emotional expressions established by psychologists which are pretty standard there,” said Patrick Ehlen, vice president of AI at Uniphore.
  • But Ehlen recognized the limitations of the technology.
  • “There’s no real objective way to measure people’s emotions,” he said. “You might smile and nod, and actually think about your vacation next week.”

-Kate Kaye (E-mail | Twitter)


Mergers and acquisitions and workforce reorganization can create a wealth of opportunities for companies seeking rapid growth, transformation and market expansion. In fact, 47% of executives say continued business mergers and acquisitions, joint ventures, and alliances are their top growth drivers in 2022. Unfortunately, nearly half of executives say acquisition and Talent retention is the biggest hurdle.

Learn more

How not to handle a breakdown

Atlassian’s Jira outage, which we covered last week in this section, is still somehow affecting hundreds of the company’s customers and potentially thousands of developers around the world. Late Tuesday, the company finally released more details about the cause and extent of the outage, proving once again that delaying bad news to corporate customers is worse than the bad news itself.

“Let me start by saying that this incident and our response time are not up to our standards, and I apologize on behalf of Atlassian,” said CTO Sri Viswanath. in a blog post on Tuesday.

He explained that last week, Atlassian engineers attempted to disable an old application that worked with Jira Service Management and Jira Software and is now fully integrated with its current services, but internal communication issues and a bad script deactivation actually caused “sites for about 400 customers”. [to be] incorrectly deleted. The incident also took away Confluence and Opsgenie, two Atlassian products that customers use to manage their own internal incident response systems.

Compounding the error is the lack of automated backup and recovery tools for an incident of this nature, which requires Atlassian engineers to manually restore affected customer data to ensure nothing is wrong. arrives at customer data that was not affected by the initial incident. incident. Expect that to change later this year.

Gergely Orosz, former Uber and Microsoft engineer, could have summed it up better:

“Outages have happened, are happening and will happen. The root cause is less important in this case. What matters is how companies react when things go wrong and how quickly they respond. And it was in speed that the company failed above all else.

—Tom Krazit (E-mail | Twitter)

Welcome to memory

European researchers believe they have found a way to advance two futuristic computing concepts – quantum computing and artificial intelligence – through the use of memristors that can simulate the neural networks used in machine learning. As detailed in IEEE Spectrumthey have developed a “quantum memristor” that works under relatively normal conditions, unlike the vacuum that many quantum computers need to maintain their state.

Memristors are electrical components that combine the characteristics of memory chips and resistors, and so they can both store and process data. The researchers demonstrated how a quantum memristor could achieve the superposition states needed to combine neural networks with quantum computing, although the experiment won’t be a real success until they can prove that such a system works with several chained quantum memristors.

—Tom Krazit (E-mail | Twitter)

Around the company

Several cybersecurity companies that work on national infrastructure systems such as pipelines and power grids, such as Claroty and Honeywell, urge the federal government to develop standards to better protect these assets against future hacking attempts.

Chipmaking Equipment Sales Soared 44% Last Year at $102.6 billion, according to Semiwith equipment sales to China outpacing the rest of the world.

Intel has tried and failed many times over the past few years to shake Nvidia’s lead in the AI ​​chip market, and The Wall Street Journal examined his latest project.

Google has developed a new technology for networking data centers that it thinks it will achieve low latency connections that current technologies can provide, which could be huge for real-time applications currently bottlenecked by the network.


ProEdge can help you perform a skills gap analysis in your organization and gain insights you can leverage to develop forward-looking plans while considering the needs of the entire business, including individuals, teams and functions. In an M&A scenario, a development program like ProEdge can also be used to uncover employee skills that were previously unused.

Learn more

Thanks for reading – see you tomorrow!

Comments are closed.