Humans and Bots: the new blended workforce

Despite robotic process automation being a growing trend, there’s still controversy surrounding the practice. While businesses continue to praise the benefits that a “blended workforce” delivers, employees are wondering how soon they will be replaced by robots. But how exactly do humans and robots coexist today? And is there any reason to sound the alarm of a “robotic apocalypse”?

In this post, we’re going to investigate some of the most common positive and negative aspects of human-robot experiences to find the answers.

Interesting fact:

Artificial intelligence: great expectations

Ever since the first mentions of artificial intelligence in Alan Turing’s “Computing Machinery and Intelligence” (1950), people have had high hopes for the positive transformations it would bring. According to a survey by PwC, 63% of respondents believe that AI will provide solutions to the most pressing issues of today. In the meantime, 46% of respondents are convinced that AI will replace the human workforce.

Consumers believe that AI will help cope with common modern problems, such as:

The blended workforce transformation

In 2016, as most companies had been moving towards a “gig economy”, Forbes’ Dan Schwabel named the “blended workforce” one of the workplace’s most dominating trends. About 93% of businesses would team up freelance workers with employees to work on projects collaboratively. Hiring on-demand workers to solve problems fast while cutting the costs associated with permanent employees would become a popular option.

In 2018, the MIT Technology Review analyzed 18 reports on the effects of automation and technology on labor which resulted in a series of divergent predictions. While some reports forecasted a one billion increase in jobs worldwide by 2030, others insisted on a loss of two billion jobs! A McKinsey 2017 report suggested 400 to 800 million jobs could be automated by 2030.

In 2020, the “blended workforce” got a revamp in the form of a human-machine partnership. A study by Honeywell Integrated suggests that over 60% of US companies are willing to invest in automation to cope with the challenges brought on by the global pandemic. Experts believe that by 2025, more businesses will be willing to invest in robotic process automation, with total expenditures expected to hit a staggering $25 billion.

Still, when it comes to human-robot interaction, the nature of the process isn’t always obvious. Let’s take a look at some examples of positive and negative experiences.

By the way, you can watch the video below to learn how to use no-code automated Bots for managing your workflows in airSlate:

Human-robot interaction: the good

GPT-3, a Bot capable of writing essays

Can a robot write a full-blown essay from scratch? There’s one that coped with the task better than an average human would! OpenAI’s Generative Pre-trained Transformer 3 (GPT-3) is known to be the most powerful natural language technology to date. It has 175 billion parameters which allow it to generate pieces of text in a variety of styles. GPT-3 is capable of generating a coherent text after being given a few simple prompts.

Here’s what it wrote in an essay to convince people that AI has no intention of destroying the human race:

Source: MIT Technology Review

Currently, GPT-3 is implemented in a vast range of cases, from auto-completion of Excel spreadsheets to NLP-based semantic web search.

Automation in pharma to develop a COVID19 treatment

Following the outbreak of COVID-19, the pharma company Takeda began recruiting patients for clinical trials of a COVID-19 treatment using automation software from UiPath. Normally, the candidates’ selection procedure and paperwork would take a few weeks. Using automation Bots significantly sped up tasks like opening files, selecting input fields, and filling out forms. The process was repeated for each prospective candidate, allowing paperwork to be processed and approved in days rather than weeks.

airSlate Bots: efficient business automation

Unlike multiple Bots geared to perform an isolated task (for instance, chatbots), airSlate Bots are rule-based automation Bots that are adaptable to nearly all kinds of workflows. Since each workflow is unique, airSlate’s capacity isn’t limited to building a flexible document or data flows.

In airSlate, all Bots are divided into Automation and Integration Bots. While the former is used for automating a particular workflow, the latter is meant to integrate it with cloud services. The result is that business processes such as contract management, onboarding requests and approvals, and patient intake, are all efficiently processed within airSlate.

Here’s how airSlate helps San Diego’s NewSchool of Architecture & Design:

With the help of airSlate, NewSchool of Architecture & Design streamlined the process of student form submission by creating required rules directly into the forms. This way, the documents are forwarded to the correct department for approval and then back to the Registrar’s Office in the form of an email that gets delivered to their Customer Management System.

Set up a fully automated no-code

workflow for any business operation

Try airSlate

Human-robot interaction: the bad

When it comes to the practical use of robotic automation and other advanced technology in daily life, there’s always room for underachievement and concern.

Amazon’s automated management

In 2018, the result of Amazon’s automated management came as a shocker. The company fired around 300 full-time employees at a single warehouse facility between August 2017 and September 2018 for failing to “meet productivity quotas.” At the time, Amazon was using an automated tracking system that rated every worker by a fixed number of items they had to process per hour. If a worker failed to meet the requirements, they would be “automatically fired”. According to Amazon’s staff, the working conditions were harsh as the robot would force them to work nonstop which eventually resulted in occupational injuries.

Bots and the “infodemic”

Social Bots are semi- or completely automated social media accounts used to manipulate public opinion. Unlike useful software automation Bots, social Bots are typically associated with malicious activities and misinformation. In 2020, Facebook, Twitter, and Reddit banned thousands of Bot accounts contributing to the COVID-19 and the 2020 US elections infodemic.

Research suggests that the social presence of Bots rapidly increases when political or economic interests are at stake:

• 82% of the top 50 most influential retweeters discussing COVID-19 were identified as Bots. Meanwhile, humans collaborating with Bots accounted for 66% in discussions around controversial pandemic-related topics.
• 71% of Twitter accounts mentioning trending US stocks, are likely to be Bots.
• Before the 2020 US presidential election, right-leaning accounts on Twitter exceeded the number of left-leaning accounts by 4 to 1 among bots and by 2 to 1 among humans. Bots have also contributed to the spread of conspiracy theories, such as QAnon and “pizzagate”.

Microsoft’s AI chatbot and racism

In 2016, Microsoft had launched a Twitter chatbot named Tay which backfired by making a bunch of racist statements learned from Twitter users. According to Microsoft’s reps, Tay was meant “to engage and entertain people where they connect with each other online through casual and playful conversation.” Sadly, Tay’s conversation didn’t go well, as the chatbot started making inflammatory and political statements in its tweets.

Microsoft later explained that the Bot used AI together with an editorial written by a team of improvisational comedians. To do the Bot justice, Tay was only repeating the words of other users. However, it was still learning from those interactions. Eventually, the Bot was “silenced” by Microsoft hours after its launch.

Source: @TayandYou/Twitter

In a nutshell

Despite the concerns associated with the use of a human-robot blended workforce, the future has already arrived. Humans need to develop mutually reinforcing, not competing, skills for interacting with technology. While AI can perform multiple tasks better than humans, it’s still humans who interpret and put the results of its work into practice.

One of the ways to gain the necessary skills needed for effective human-robot collaboration is by completing Automation Courses with the airSlate Academy.