Artificial Intelligence is reshaping the workplace as well as society, but no news there. Also, no need to worry about robots stealing our jobs, not yet at least.
Researchers have found that, whilst some jobs might be replaced by AI automation — while creating opportunities for new, previously unforeseen professions to arise — the key change brought by widespread adoption of AI projects will be a close collaboration between humans and machines.
“The best results will come from humans supported by intelligent machines”, writes James Timbie, Distinguished Visiting Fellow at Stanford University, in Beyond Disruption: Technology’s Challenge to Governance.
“In the workplace of the near future…the machines [will do] the computational work they do best, augmenting the humans who see the big picture and have interpersonal skills.”
AI adoption will impact businesses across the board — in his book, Timbie refers to “a combination of a doctor and a machine, a teacher and a machine, and so on”.
Big data, big problems
As an AI company that operates in business intelligence and analytics, we like to think of “the workplace of the near future” in terms of a new type of data interaction that empowers — indeed, augments — workers to make the best possible decisions based on data they have, but lack the skills, time and tools to analyze.
You know the spiel: the amount of data we produce on a daily basis is staggering, and bound to get larger. Analyst firm Forrester estimates that over two thirds of the data enterprises hold at present goes unused: big data has created some serious storage issues, widening the gap between data stored and data put to use. To oversimplify, this is due to two main factors:
- Lack of specialist skills. The shortage of people with data science skills amounts to over 150,000 in the US alone, according to a 2018 LinkedIn study.
- Time. Finding data that is scattered in different sources, often in different formats, organizing it, analyzing it, and crunching it into insights takes time.
Whilst big picture, strategic decisions may be worth the bandwidth and wait, every-day operational decisions constantly miss out on the value and context data can add.
This is where AI can help.
A virtual advisor can augment workers’ ability to make informed decisions by doing the legwork of bringing all scattered data to one place, analyzing it in seconds as opposed to days, and turning it into insights that are immediately actionable.
Ok — but what exactly is a virtual advisor?
Consider Timbie’s statement about the best results coming from “humans supported by intelligent machines”. In this scenario, humans and machines work together as colleagues, and that is precisely the role of a virtual advisor.
To oversimplify again, in the context of business intelligence and analytics, a virtual advisor is a smart colleague who removes the time and skills barriers that prevent operational workers across industries from making informed data-driven decisions. Which still begs two questions: what kind of decisions? And, how do virtual advisors actually do that? Let’s break it down:
📱Mobility, mobility, mobility
The idea of an intelligent virtual co-worker available to users via their smartphones anytime, anywhere, and capable of simplifying tasks as complex as organizing, analyzing and extrapolating data underpinned the development of our very own virtual advisor for data intelligence, crystal.
Mobility is a key element here, but again that is not surprising as (a) remote work is on the rise, along with the development of new remote working tools, and (b) some workers — sales reps, for example — essentially need to make decisions on the go.
💫 Crunching data at the speed of light with fast data retrieval
Fast data retrieval is the technology pillar allowing crystal to slice and dice vast amounts of data in real time, and to provide instant answers to users’ queries, empowering them to operationalize data in a way that would have otherwise required time and specific resources.
We like to call it data democratization: just as CEOs would ask a team of data scientists to crunch some numbers to support an important roadmap choice, crystal users can turn to AI for help with daily decisions.
In an industry 4.0 context, for example, our award-winning work with energy giant Enel helped the company save significant time and resources when monitoring the performance of its Global Thermal Generation Division’s thermoelectric plants.
Enel estimated that, due to the vast amount of data the company gathers, 80% of its workforce’s time was spent finding, assessing and analyzing data. crystal was able to cut that time to seconds, leaving Enel workers plenty of time to prevent or react to any power plant anomalies.
📢 Making data talk with conversational AI
Simplifying access to data takes on a new meaning when mobility meets conversational AI, allowing users to query their virtual advisor using their voice, and to receive insights in natural language in response.
We like to say that crystal makes data talk, removing another layer of complexity when working with business data: an elaborate graph or packed spreadsheet may not be the most user-friendly thing to read and digest, especially whilst on the go. Receiving insights via voice, sort of like asking a colleague’s opinion, is much simpler.
Our work with leading insurance company Allianz allowed the company’s insurance agents to access data such as income monitoring, performance of a particular product, and client profiles on the go, simply using a smartphone and their voice.
Conversational AI is a key component when it comes to business innovation, as we are in the middle of a type-to-voice technology shift: as voice-controlled devices proliferate, we will progressively type less. US advisory firm the Future Today Institute predicts that by the end of 2020 half of the interactions we’ll have with computers will be using our voice and not a keyboard.
💪🏻 Connecting the dots with machine learning
Machine learning and deep learning algorithms allow AI to progressively learn from data and add context. Making connections between data, virtual advisors can therefore enrich the answers to users’ queries with insights and even estimates and predictions, as opposed to simply providing raw numbers. Providing recommendations is after all an advisor’s defining trait.