AI at work – Recruiting, Chatbots and Trading

November 3, 2020

The use of AI in the workplace is quickly becoming commonplace. Sure, we haven’t got robot slaves bringing around the biscuit tin quite yet, but the productivity increase due to AI automation cannot be understated. AI’s deft handling of repetitive tasks helps a company cut back on unnecessary expenditure. And it gives more time to human employees to do more significant, creative work.

Here we will explore the three areas where AI is making the biggest splash:

1 – Recruiting

The recruitment process in 2019 would appear totally alien as little as 20 years ago. You’ll find the majority of jobs in the UK listed online using sites such as Indeed and LinkedIn. For many companies, this means receiving a huge volume of applications. Fewer and fewer of these applications are seeing actually human inspection. Instead automated AI recruiters pick out desirable traits, level of education, quality of written English etc.

There are a number of risks in going down this path, however. Teething problems mean that a company’s incorporation of AI can lead to some initial problems. Poor matching can waste time and money, while letting potentially perfect applicants slip through your fingers. Simultaneously regulation is tightening its hands around AI recruitment processes. These technologies, while using closely guarded proprietary algorithms, are not granted any regulatory oversight. Discriminatory behavior can be learned by AIs. The Society for Human Resource Management suggests that “if a company’s highest performers historically have been identified as white males between 30 and 40 years old, because those individuals were frequently promoted into next-level jobs, that bias can inadvertently become build into algorithms that learn from talent management patterns.”

Despite these issues, AI automation continues to forge its way into the recruitment spotlight. Why is this? 41% of employers estimated a single bad hire costs $25K, and 25% put the figure at $50K or more. Moreso there is a 10-year record high voluntary quit rate of 62%. It is clear that getting the right person for the job can be a headache.

Recruiting company AllyO has been making stirs in the industry with its chatbot, and talent re-discovery tools. AllyO can screen candidates for various necessary traits, schedule interviews, and give other assistance. However, most impressive is its ability to seek out candidates from a company’s list of talent and prior applicants. The chatbot can asks old candidates about their current work situation, and suggest that they put their name forward for new positions.

This kind of proactive, scaleable recruitment solution can sort through hundreds of applicants or potential applicants simultaneously, helping to find great matches for any job.

2 – Customer Service Chatbots

Although becoming increasingly present in recruiting, chatbots are almost ubiquitous in customer service. They are frequently on the front lines of customer interactions. They help to support an increasingly large demand for instant 24/7 support, and cutting down on customer support expenditure. The benefits do not end there as integration with Customer Relationship Management software can create an end to end documentation of customer issues and history, enhancing business strategy and helping customers from losing their minds.

Chatbots are available as instant messaging services, but also as live voice chat bots. Long gone are the days of robotic women saying “I’m sorry, I didn’t quite get that”. The new generation of voice chatbots use cutting edge NLG to achieve an incredible level of accuracy. More than just translating voice to text, they are able to decipher meaning from these words and match it to possible assistance options. Traditionally, a customer vocally chooses from a list of predetermined pathways. If this fails, at it frequently does, the customer is then transferred to a customer service phone operator. Now, on the other hand, you will simply be asked to describe the issue that you’re having in your own words. The AI conversation can continue from there to help you solve the issue, or you can be transferred to a specific member of the support team.

General chatbots, the kind we used to converse with on school computers, create stilted and nonsensical outputs. This is a result of the huge scope of, well, everything! In customer services the scope can be narrowly defined to a company’s exact operations. In this environment, chatbots do very well indeed, frequently becoming more useful than their human counterparts while saving money in the process.

3 – Trading

AI in financial markets has been around for longer than you may think. As early as the 1980s, we’ve been using rule based AI systems for trading. The complexity of financial markets means vast datasets – the perfect environment for AI to thrive.

The financial services sector has not stagnated, and has been quick off the mark when it comes to adopting AI advancements. Rules based systems were a rough guide at best, and more advanced, algorithmic, machine learning techniques have generated greater returns on investments, and reduce risk.

AI’s are able to view an industry, sector, or index in its entirety, something human traders would have great difficulty doing. This opens up a new world of trading signals, trends, and risk markers which are only accessible through AI systems.

J.P Morgan estimated in 2017 that only around 10% of trading is done in the traditional way (systematic, human risk/reward assessment). 60% is described as passive or quantitative trading. This is the category which AI trading bots fall under. The lines do, of course, blur in places. Many traders, using a more hands on method, will augment their decision making with curated news feeds, automated market analytics, and some tools which border on AI.

At Agrud Technologies we help companies to enhance their decision making through an array of tools.  This includes News Sentiment Analysis, content curation, and natural language generation, as well as financial portfolio tools. These kinds of tools are frequently integrated into credit risk management teams, using AI analysis to turn soft data (news, blogs etc) into hard data.

Agrud Technologies Dashboard Ai Credit Risk Sentiment Analysis
Agrud Technologies Dashboard

Apart from these big three, there are a couple of honorable mentions that also deserve a bit of attention.

NLG Journalism

Natural Language Generation in journalism has been making waves for the past few years. In 2018, a Natural Language Generating AI named Tobi was responsible for reporting on the results of Swiss referendums for 2,222 municipalities. Similarly, in 2017, Press Association won a £621,000 grant to start producing automated regional news. They partnered up with Urbs Media and are working towards producing up to 30,000 local news stories every month. Despite these impressive use cases, the scope is currently limited to hard data reporting. In the future it is likely that NLG will be able to interpret more nuanced events.

AI in Regulatory Compliance

Sure, regulatory AI isn’t very glamorous, but using AI to empower regulatory compliance automation could prevent huge fines. The money involved creates a compelling case for AI Regtech solutions. But the savings in time and manpower are also considerable. A Duff & Phelps survey suggests that we will see compliance costs in the financial sector double by 2022 – a concerning figure considering the current spend for compliance and regulatory obligations is already $270+ billion per year. Because of this, an increasing number of companies are putting AI to work in managing regulatory changes, preventing them from making costly mistakes.

AI is finding space for itself in most industries and sectors. Although the AI buzz has been present for a long time (since well before the first screening of ‘The Terminator’) AI companies are now transforming primitive, rule based AI into deep learning, futuristic visions such as IBM’s Watson. With the capability to save time and money, protect against risk, generate high quality analytics, and help deliver superior customer experiences, it is no wonder that AI is quickly becoming the greatest force for disruption across a multitude of sectors.