What are AI chatbots?

Be the first to comment

What are AI chatbots?

Contributed

This content is contributed or sourced from third parties but has been subject to Finextra editorial review.

The hottest trend in tech and financial services has been artificial intelligence (AI) and its infinite potential to revolutionise and transform the industry to make it more efficient, reliable, secure, and accessible.

However, as AI grows in popularity, fear of the sentient AI also lurks in the back of many minds. Not exactly the apocalyptic Terminator anxiety, or even the I, Robot of it all (those fears are for future humans!) but simply the everyday Joe’s worry about AI replacing human labour, taking jobs, and leaving the people it is meant to aid behind.  

With the recent news from Klarna on how their AI chatbot has the capacity to easily take on the workload of 700 full-time humans, the technology is formidable, but its impact has yet to be considered. What are chatbots exactly, and how do they impact us?

The evolution of the chatbot

Chatbots evolved from FAQ programs using keywords and phrases to support users, to being able to process complex questions and deliver equally detailed answers. The advancement from key words to actual ability to comprehend and respond to voices and description is the result of natural language processing (NLP), machine learning, and deep learning.

AI chatbots expedite the customer service process rapidly, now that they are able to recognise human voices, with their various accents, inflections, and intonations. You may realise that often when you phone up your bank, they have a robo-advisor listen to your voice and direct you to a service accordingly. Deep learning and natural language understanding (NLU) is what makes chatbots able to become more advanced the more they are used and learn over time.

While ‘chatbot’ is an overarching term, ‘AI chatbots’ refer to those that use AI technology, and ‘virtual agents’ are developed AI chatbots that use AI and robotic process automation to offer the user further services related to their initial query. ‘Generative AI chatbots’ are the next step, that can understand more complex questions and use a variety of mediums to create content in their response. GenAI chatbots are able to translate, predict, spot, and build content without human interaction.

Now AI agents are a part of everyday life, you have your pick from Siris, Alexas, Geminis, and of course the pinnacle of AI assistants, ChatGPT. These guys are your customer service AIs, intended for support and customer care.

Chatbots in financial services

The benefits of AI chatbots are revolutionary, particularly from a customer experience and data analysis standpoint. AI chatbots are able to personalise shopping experiences on e-commerce platforms by suggesting related products that a consumer might be interested in and facilitating order tracking.

On the back-end of banking, AI chatbots can remove human error completely and expedite the data collection and analysis process. Nowadays, more companies are using AI chatbots for financial advice and to detect fraud, which is becoming increasingly prevalent as cybercriminals become more sophisticated.

Ethical concerns

The challenges to implementing AI chatbots and AI in general, are concerns for privacy and data security. AI chatbots live off data, and the more data they consume, the more they are able to support the userbase. However, in financial services and e-commerce markets there is sensitive data at stake, that consumers often are worried about putting into a system which can hold and share information.

Additionally, AI bias is prevalent depending on those who develop the software. Generating AI software that does not include unconscious bias requires a wealth of information and multiple rounds of checks.

A significant concern is AI replacing jobs, which is already in practice for companies like Google, Salesforce, and MSN. Google and Salesforce reportedly conducted major job cuts to replace workers in customer service and sales with AI. MSN now uses AI for media content instead of journalists. According to a survey from The Conference Board, 50% of CEOs believe that AI will lead to labour displacements in their company.

Future trends

While AI can displace employees, there is also space for AI to create jobs for AI specialists, developers, data curators, and more.

The next step for AI chatbots is for GenAI chatbots to become complex enough to measure all sorts of natural interactions, possibly using biometric technology which is already on the rise in the banking sector. GenAI chatbots could potentially possess emotional intelligence (I’d like to think adjacent to Wall-E or C3PO).

What’s next for AI chatbot technology is for businesses in the industry to focus on responsible deployment that maintain customer security and privacy, and to prevent bias within software. Additionally, AI regulation is stepping up globally – with the 2023 EU AI Act, plans for an AI Bill of Rights in the US, and an upcoming UK bill - that will keep our budding robo-pals in check.

Channels

Comments: (0)

/devops Long Reads

Sehrish Alikhan

Sehrish Alikhan Reporter at Finextra

What are AI chatbots?

/devops

Dominique Dierks

Dominique Dierks Content Manager at Finextra

Why data is the backbone of predictive AI

/devops /ai

Duncan Cooper

Duncan Cooper Chief Data Officer, Asset Servicing at Northern Trust

What are the key data trends for 2024?

/devops

Contributed

This content is contributed or sourced from third parties but has been subject to Finextra editorial review.