Talkbot Basics

14
Nov
Talkbot Basics  ·  Voice AI Technology
AGI Holds the Power to Rewrite All Business Paradigms, from Startups to Tech Giants
Jennifer Zhang

Jennifer Zhang

CEO & Co-Founder

Envision a world where the multitude of apps becomes obsolete, replaced by a single AI agent that can schedule your appointments, make calls, order food, and handle most paperwork. This is not a distant dream but an emerging reality that will soon become an integral part of our everyday lives. 

During an in-depth seminar held by WIZ.AI in Singapore on November 7, 2023, Dr. Hongjiang Zhang, a Fellow of IEEE and ACM and a 2022 electee to the National Academy of Engineering (US), reassured that the AI should not be feared for job displacement. Instead, the forthcoming Artificial General Intelligence (AGI) era is poised to create unprecedented job roles and opportunities. Both tech giants and startups are well-placed to enter this arena and carve their niche.
 
In the wake of ChatGPT, the software industry is facing a seismic shift that will inevitably reshape everyday life and work, according to Dr. Zhang. He also highlighted that while the AGI era will likely affect most jobs, it will also create unprecedented career opportunities. The key, he suggests, is leveraging a large language model (LLM) to enhance personal efficiency.
 
Dr. Zhang discussed the strides made by tech giants like OpenAI, Microsoft, and Google in LLM research and applications. He urged us to view ChatGPT as a potential ‘foundation model,’ which could pave the way for AI capable of communicating in natural language, images, and video. He pointed out the past focus on vertical, industry-specific AI models that lacked interoperability, a limitation shattered by the foundation model concept, heralding vast commercial possibilities.
 
Citing OpenAI CEO Sam Altman, Dr. Zhang equated the rise of AI to a fourth technological revolution, succeeding those of agriculture, industry, and computing. He referenced a U.S. study on AI model’s impact on labor market indicating that 80% of the workforce could have at least 10% of their tasks impacted by LLMs, affecting all income levels, with high-earners potentially facing more significant changes.
 
What Will Happen to the Rest? Unemployment or Idleness? Dr. Zhang suggested that the AGI era will usher in entirely new business models and companies born from AGI technology. As AI models mature, their applications diversify. Consumer-facing tasks like document writing, chatbots, code generation, and meeting transcriptions are already being automated. The future could see the creation of a new platform to replace app stores, where all needs are met by AI plugins, leading to significant shifts in B2B business models.
 
In the realm of Software as a Service (SaaS), customer support could be the vanguard of sectors transformed by AI. A survey of international corporations revealed that more than 57% intend to employ ChatGPT for customer support. Hence, the most sought-after B2B companies might soon be those facilitating services for the entire AI ecosystem.
 
Take WIZ.AI as an example, a rapidly growing voice AI provider in Asia, pioneering LLM application with its service framework, offering tailored, multilingual, and multimodal tools for various industries. WIZ.AI has notably used LLM tools to achieve over 95% accuracy in telephone orders, halving costs, and setting the stage for widespread industry adoption.

06
Oct
Featured Articles  ·  Main Posts  ·  Talkbot Basics  ·  Voice AI Technology
Voicebots Vs LLM agents: Navigating the Future of AI for Your Business

Artificial Intelligence (AI) is revolutionizing how businesses operate. For companies, the path is full of potential yet accompanied by complexities. In this article, we would like to guide you through this intricate AI landscape, by unraveling two key AI technologies, the established voicebot and the rising LLM agents.

What is a voicebot?

When you receive a phone call reminding you about an upcoming appointment or pending repayment, you could already be talking to a voicebot. Voicebots utilize Automatic Speech Recognition (ASR), Natural Language Processing (NLP) and Text to Speech (TTS) technologies. They can really understand how people talk and stimulate human-like interactions based on customer reactions.

A common use case for voicebots is 24/7 customer service. They can handle common inquiries from inbound calls at any time, so human agents can focus on more complex issues.

Voicebot also excels at high volume outbound calls for reminder, promotion and leads filtering. They are effective to help businesses to boost their customer and revenue growth. Voicebot can make customer outreach calls on a massive scale to help remind all relevant customers, or quickly filter and qualify prospects in the entire CRM contact list. With voicebots handling initial outreach, human agents can follow up on qualified, sales-ready leads.

What is an LLM agent

As we introduced in our previous insight article, an LLM agent is a combination of a large language model with three other key components for planning, memory and tool use. We can also refer to an LLM agent as a LLM powered autonomous agent system, in which the LLM functions as the brain.

LLMs have unveiled their powerful Natural Language Understanding (NLU) capabilities. But LLMs alone are like brains in vats, and they can not realize a wide array of real-world applications. To unlock the full potential of LLMs, we need to build systems that can acquire and apply knowledge to solve practical problems. That’s where LLM powered autonomous agent system or LLM agents comes in. LLM agents can perceive, reason and act like our copilot, beyond just conversation capabilities. In theory, when giving LLM agents any general objective and a robust enough agent system, they can plan for task completion step by step. They are also capable of calling available agents or tools for help when necessary.

Comparing voicebots and LLM agents

Voicebots adhere to rigid pre-determined dialogue flows. They only understand the expressions and words they are trained for. For instance, a seemingly simple use case like transferring a hospital patient’s call from the main line to different department extensions could require a matrix of dialogue flow. This could take weeks for conversation designers to develop.

In contrast, LLM agents possess a new edge in their potent Natural Language Understanding (NLU) capabilities. They can understand diverse human expressions, comprehend in context during real-time conversations. If your query includes something LLM agents don’t understand, they can continue the dialogue until they achieved clarity. This closely resembles how a human would ask follow-up questions to understand better.

Furthermore, while voicebots can perform what they are trained for efficiently, they lack autonomous planning or decision-making capabilities. Nor can they adapt to dynamic external environments.

In comparison, LLM agents are capable of self-learning new knowledge across languages and knowledge domains. They possess both short-term memory from in-context learning, as well as long-term memory from search and retrieval. They also can perform reasoning and planning, breaking down large and complex tasks into smaller, manageable steps.

How would a refund be processed differently

Let’s consider a customer refund request scenario to understand how a voicebot and an LLM agent would act differently.

A voicebot can understand all expressions and keywords meaning refund if trained for them. It can read out the refund process from a FAQ list when a refund is requested. However, a voicebot won’t be able to handle queries beyond the FAQ list.

On the other hand, an LLM agent can call different tools and leverage industry best practices to resolve the issues. Moreover, it is capable of continuous learning and optimization, acquiring new knowledge and best practices overtime. When a client requests a refund, a LLM agent can ask for the purchase ID, then call relevant agents or tools (i.e. logistics tracking sheet) to check delivery status. They can even soothe the client, suggesting waiting for a couple of days before processing the refund. Meanwhile, LLM agents would analyze order time, and assess whether the no-fault refund period has expired, or check the customer’s credibility status to decide if a refund is eligible.

Beyond customer engagement

Looking ahead, we expect LLM agents to go far beyond inbound and outbound customer engagement. They will fully integrate with the enterprise software system, and become the core command center of the entire system. By then, LLMs will be industry experts and decision makers. They would understand domain-specific enterprise knowledge and dynamically call different tools to automate task completions. The agents or tools being called by LLMs can vary, including common office systems such as CRM, ERP, OA and PMS etc. We envision LLM agents to be versatile copilots, managing workflows alongside employees and assisting customers.

Choosing the right AI solution for you

Despite the promising future of LLM powered autonomous agent systems or LLM agents, their development is still in early stages. LLM agents can alleviate the burden of training and setting up intricate dialogue flows. They have the capability to learn and optimize with minimal human intervention. However, the training of such LLM agents demands significant computational resources and GPU capacity. Therefore, LLM agents might not always be the most cost-effective solution for your business.

Voicebots, conversely, serve as robust business copilots, and are capable of driving impressive results. They enhance customer satisfaction by providing round-the-clock services, augment payment rates through automated reminder calls at scale, and efficiently qualify leads for human agents to follow up, among other tasks.

Consequently, we recommend businesses leverage voicebots for relatively straightforward applications like sending reminder calls, executing preliminary lead filtering, or addressing routine inquiries. Businesses should consider tapping into the power of LLM agents when they necessitate complex interactions, dynamic reasoning, industry-specific knowledge, or integration with a variety of enterprise systems.

The world of artificial intelligence offers a broad spectrum of solutions for businesses. Whether it’s a voicebot or an LLM agent, the choice depends on your specific needs and resources. The key to success lies in understanding these AI systems, and choosing the right AI assistant to navigate your business towards a prosperous future.

Reference

Zhiheng Xi et al. (2023). The Rise and Potential of Large Language Model Based Agents: A Survey. Retrieved from: https://github.com/WooooDyy/LLM-Agent-Paper-List


20
Sep
Featured Articles  ·  Main Posts  ·  Talkbot Basics  ·  Voice AI Technology
How Voice AI is Rapidly Driving Customer Growth

As digital transformation reshapes industries, companies are increasingly leveraging artificial intelligence (AI) to reimage their business operations. Among various AI technologies, voice AI-powered omnichannel engagement solution stands out as a powerful tool, as it enables businesses to automate huge amount of phone conversations while delivering hyper-personalized communication based on customer reactions. Today’s customers prefer this kind of direct human connection to address their queries and concerns in real time. In this article, we’ll delve into how Voice AI, particularly WIZ.AI’s TalkGPT, is revolutionizing customer engagement, leading to rapid customer growth for businesses.

The power of voice AI for customer growth

Nothing beats a human voice conversation when it comes to engaging customers and driving growth. Telemarketing has been a classic yet popular marketing choices for many businesses. It delivers instant interactivity and feedback that emails, ads, messages, or website visits simply can’t match. Customers tend to give phone conversations their full attention. They listen to brands’ offerings, ask questions, and provide valuable intent signals, allowing for personalized follow-ups. It’s these real-time insights into customer needs that enable impactful conversations and drive greater customer growth.

Voice AI solutions can mimic this kind of natural and real-time human interaction well. Utilizing Automatic Speech Recognition (ASR), Natural Language Processing (NLP) and Text to Speech (TTS), voice bots can really understand how people talk and stimulate human-like interactions based on customer reactions. They respond appropriately to questions and objections, capturing key insights and identifying high value leads for human agents to follow up. More importantly, voice AI allows companies to engage audiences on a massive scale, i.e. millions of calls in a few hours. The bots can work around the clock to spread out time-sensitive promotional offers and filter high-potential leads.

This perfect balance of automation and personalization empowers businesses to scale their outreach and create human connections that is critical for gaining trust and buy-in.

Introducing TalkGPT: a game changer for customer engagement

TalkGPT is a self-service generative AI-powered talkbot by WIZ.AI. It is a game-changer in proactive outbound customer engagement. With zero set-up cost and quick deployment, businesses can build and launch customized Talkbot campaigns that provide human-like and automated phone conversations at scale, in just a few minutes.

Tired of relying on professional support for every script tweak? With TalkGPT, you are in complete control. Quickly tailor conversations to your branded voices and optimize results, no code or AI experience needed. Such rapid adaptability is crucial in situations that require constant iterations in marketing messages or during A/B testings to identify the most effective approach.

High scalability with human touch

The power of TalkGPT: real-life use cases

TalkGPT provides conversation templates based on known best practices. Businesses can leverage these handy templates and quickly draft campaign messages. Companies can also add a FAQs list. With the list added, TalkGPT can smartly pull answers from FAQs during live conversations, when asked relevant questions.

Based on our conversations with clients, we found TalkGPT is particularly effective in two main scenarios: running promotional campaigns and filtering leads.

Scalable promotion campaigns

Capitalizing on flash sales and limited-time offers is key for boosting conversions. But ensuring your time-sensitive promotions reach customers quickly at scale remains a challenge. TalkGPT delivers targeted promotional messages on scale(up to 1 million in an hour), based on purchase history and CRM data. Real-time intent collection allows for tailored follow-ups to interested prospects.

The bot also handles hesitant customers well. By probing why these people are undecided, TalkGPT uncovers key insights to further define them with a “neutral” or “not interested” tag. The insights can be leveraged to refine future campaign targeting and messaging.

Conversation script between a TalkGPT bot and a human customer in more detail. The car insurance company also added FAQs for the bot to handle customer queries.

Lead filtering

TalkGPT can be a powerful ally for human agents when selling high-value products such as property or insurance. By making initial calls and asking qualification questions, TalkGPT can identify high-value customers and pass them to human agents for follow-up via calls or WhatsApp.

Overwhelmed by too many leads? TalkGPT provides fast, intelligent lead filtering to focus on what matters. With advanced natural language capabilities, our bots conduct personalized conversations to qualify prospects in real-time. They can engage up to 1 million leads in an hour, capturing key details like budget and transaction intent to instantly categorize hot, warm and cold leads. This takes the manual grunt work out of lead analysis while reactivating warm leads. With voice AI handling initial outreach, human agents can follow up on qualified, sales-ready leads.

Language support and data analytics

TalkGPT offers multilingual support, including English, Bahasa, Taglish, Thai, Australian English and more to come, making it a versatile tool for local enterprises as well as businesses operating in multilingual markets.

The product isn’t just a communication tool; it’s also a data powerhouse. It offers insights into call data, customer behaviors and intentions. Businesses can find out insights such as how many people picked up the calls, how many showed interest, and when customers are most available for calls.

Elevate TalkGPT’s outreach impact on WhatsApp

Seamlessly blending AI conversational ability with instant messaging, TalkGPT’s integration with WhatsApp provides businesses with an omnichannel solution to engage and convert modern customers. Here are a few ways businesses can integrate TalkGPT with WhatsApp to offer nuanced, personalized interactions on a large scale.

Trackable campaigns for maximizing conversions

TalkGPT calls allow interactive introduction of brand offerings, such as membership benefits. WhatsApp messaging then delivers the direct membership registration link. When a potential customer clicks the link to register, this action will be calculated as a successful conversion. This end-to-end experience creates a seamless journey for customers, from initial awareness to final conversion, thereby optimizing campaign performance.

On-demand details via WhatsApp

For complex products like credit cards and investment plans, TalkGPT provides a consultative overview over phone calls. Customers can then request full details sent via WhatsApp for later review. This caters to different consumers’ preferences while capturing every potential lead.

Automated information collection

Customers can ask queries during a TalkGPT call. Subsequent follow-up will then continue on WhatsApp. Questions and responses on WhatsApp are automatically compiled into an organized sheet, ready for businesses to review.

Act now for our time-limited offer

You’ve seen the power of our voice AI-powered omnichannel engagement solution to transform customer engagement. Starting today through September 30th, we’re providing 500 minutes of FREE calls plus a no-obligation free trial of our full service!

Don’t wait – this unprecedented offer does not last long. Click here to register immediately and unlock voice AI capabilities that will revolutionize your businesses.


16
Aug
Featured Articles  ·  Talkbot Basics  ·  Voice AI Technology
Reimagining AI Governance: AI as Co-Pilot

As we delve deeper into the realm of AI, a growing complexity of challenges and risks are becoming more lucid. In a recent Guardian’s report, AI could even eavesdrop typing sound and identify which keys people pressed for passwords! How can we ensure responsible use of AI when it is becoming more powerful and ubiquitous? In this article, we are trying to reimagine AI governance, and propose AI can play a more prominent role and become our co-pilot to pave the way for trustworthy AI.

The pressing need for ethical and trustworthy AI

There are manifold risks associated with AI or generative AI to be specific. To list a few most commonly addressed, generative AI can produce inaccurate or even fabricated responses. It’s also struggling with algorithmic bias that reflects and reinforces the existing bias in society. For example, we may all heard about some AI recruitment tools that can disadvantage female candidates, or certain facial detection that works more accurately on Caucasian faces. In addition, AI-generated works are faced with the challenge of violating copyrights and data privacy. And more seriously, generative AI’s widespread availability substantially escalates risks of it being used maliciously for cyber attacks and fraud.

To mitigate and prevent risks, governments and AI industry leaders around the world are rolling out AI governance initiatives and policies. In June, EU made the first move and carried out AI Act, which is deemed as the world’s first comprehensive AI law. In July, seven leading AI companies, including Amazon, OpenAI and Meta, followed suits and introduced voluntary safeguards at the White House. And in the upcoming fall, we are expecting the new international standard ISO/IEC 42001 on AI Management Systems to be published.

Here in Singapore, where WIZ.AI is headquartered, the government takes a balanced approach to facilitate innovation and safeguard consumer interests. The government regularly facilitates dialogues and collaborations among the public sector, industry and academia. In a recent discussion where WIZ.AI participated, we also exemplified our commitment to responsible use of AI. Our company is ISO 27001 certified, which is a globally recognised standard for information security management systems (ISMS). This certification demonstrates our commitment to the highest levels of data security and privacy. In addition, we also certified by SOC2 Type II Report, an independent audit report that provides assurance on the controls of a service organization. It evaluates the design and operating effectiveness of the organization’s controls related to security, availability, processing integrity, confidentiality, and privacy.

Responsible use of AI: current practices

Besides broader initiatives mentioned, currently there are diverse AI governance practices from companies developing and deploying AI as well.

  1. Incorporate guardrails:
  • Develop compliance processes for risk assessment and review
  • Retain proprietary data within secure or localized infrastructures, under direct monitoring of the organization
  • Grant system users with different roles and data access
  • Leverage privacy enhancing technologies (PETs) to derive insights from consumer datasets, safeguarding personal data confidentiality
  • Intelligence collection about ongoing breaches and threats
  • Eliminate toxic data that contains hate, abuse, profanity content, as well as those with private information and license constraints, during the data filtering stage
  • Testing systems for AIGC tools’ safety, explainability as well as their robustness against security attacks
  1. Uplift transparency:
  • Document the data pile used for training, as well as how the model is trained and validated, increasing explainability in AI-decision making
  • Label content that’s generated by AI, or clearly inform users that they are interacting with AI systems
  1. Human validation:
  • Loop humans to validate AI outputs before public distribution
  • Create feedback loop for users to identify issues with AI system’s functionality and behaviors
  • Ongoing monitoring by humans to detect unexpected behaviors, flaws or vulnerabilities after model deployment

AI as co-pilot in AI governance

After examining current initiatives and practices for responsible use of AI, we are wondering if we are being a bit too human-centric sometimes, and overlook the role AI can play in its own governance?

For starters, Generative AI is already assisting with risk mitigation. Companies have been utilizing generative AI to “red team” AI solutions and generating edge cases to test system’s robustness. There are real-life examples with generative AI goes beyond laboratory testings too. China’s Ant Group introduced an LLMs-enabled AI assistant for anti-fraud education, resulted in a 10% reduction in reported fraud cases. Waabi, a Toronto-based start-up uses generative AI to automate vehicle safety. This helps the company operate at only 5% of the previous cost. WIZ.AI leverages LLMs to detect sensitive content for better compliance and help maintain sensitive content errors at <5% level.

Generative AI already takes up a role as an assistant mitigating risks at a lower cost, what if it takes a more proactive role and becomes a co-pilot to achieve responsible AI more efficiently?

At WIZ.AI, we decided to test the above hypothesis. We trained one LLM to monitor the other LLM for bias, violence, and profanity content in workplace chats. The monitoring LLM successfully flagged many issues based on the given prompts, without further human involvement. However, we still haven’t fully realized the co-pilot status because there are still loopholes that need human to check the root causes and tweak the prompts accordingly, and there are even cases where human cannot yet identify the root cause that leading to AI’s obvious false judgements.

With the further exploration and development of our LLM technology though, we can envision a future where human-AI co-pilot in AI governance can be achieved, when humans can understand generative AI’s decision making process on a granular level. By then, the well-trained AI models will “tutor” other models on workplace’s “responsible code of conduct” independently with zero mistake. Humans will only need to validate the learning results without too much involvement in tedious work, such as fixing errors, digging the roots causes, and shift attention to more important work.

Forward-looking of a universal responsible code of conduct

Take an even more forward-looking perspective, a universal “responsible code of conduct” module can be developed and by-default plugged into every newly launched LLM or their much more intelligent “offsprings”. There can even be specific versions by region, by country or by individual organizations building on top of the foundational module. This process ensures AI systems are “aware” of their responsibilities and ethical guidelines upfront and ultimately liberates human labor. Naturally, humans will still be in the loop, judging AIs’ “performance”after their onboarding, reviewing and iterating the “responsible code of conduct” module on an ongoing basis.

Looking ahead, we see a world where human-AI collaboration permeates every possible aspect, ushering in an era of ideal co-existence between humanity and technology.


Reference

2023 World Artificial Intelligence Conference, Trustworthy AI Forum. Retrieved from: https://36kr.com/live/2332158061807241
IBM Research. (2023). Generative AI for business. Retrieved from: https://www.youtube.com/watch?v=FrDnPTPgEmk
Speech by Minister of Communications and Information, Mrs Josephine Teo, at the Opening of the Personal Data Protection Week on 18 July 2023. Retrieved from: https://www.mci.gov.sg/pressroom/news-and-stories/pressroom/2023/7/speech-by-minister-josephine-teo-at-the-opening-of-the-personal-data-protection-week-on-18-july-2023
Brand Smith, Microsoft. (2023). Advancing AI governance in Europe and internationally. Retrieved from: https://blogs.microsoft.com/eupolicy/2023/06/29/advancing-ai-governance-europe-brad-smith/
Guardian. (2023). AI can identify passwords by sound of keys being pressed, study suggests. Retrieved from: https://www.theguardian.com/technology/2023/aug/08/ai-could-identify-passwords-by-sound-of-keys-being-pressed-study-suggests?CMP=fb_a-technology_b-gdntech
Saritha Rai. Bloomberg. (2023). Startup Sees AI as a Safer Way to Train AI in Cars. Retrieved from: https://www.bloomberg.com/news/newsletters/2023-08-08/raquel-urtasun-startup-waabi-trains-autonomous-vehicles-with-ai


12
Jun
Talkbot Basics  ·  Voice AI Technology
The Southeast Asian unicorn – and why we need more of them
Jennifer Zhang

Jennifer Zhang

CEO & Co-Founder

I’ve had the privilege of participating in a panel discussion at the ATxSG Summit, on the state of Southeast Asia’s startup ecosystem in early June 2023. The session posed a number of questions that got me thinking about many things.

Southeast Asia is quickly becoming a hotbed for innovation and entrepreneurship. With a population of over 650 million people, a growing middle class, and a tech-savvy youth demographic, Southeast Asia is ripe for startups to flourish. The region has seen a surge of unicorn companies in recent years. Below are some thoughts that were spurred by the discussion.

What makes a unicorn?

As is often seen in scrappy startups, successful unicorns solve real-world problems for ordinary people in tech-smart ways. They understand what technology can do and push the limits of possibility, disrupting traditional ways of doing things.

One of the distinguishing features of Southeast Asian unicorns is their focus on emerging markets, where they see untapped potential and a growing middle class. There is also a strong product-market fit, which is an essential ingredient for wide adoption. Unicorns also build adaptable, scalable business models that can handle rapid expansion across different markets. They are also exceptionally good at regional adaptation, understanding different cultures, languages, and preferences in the region.

Southeast Asian unicorns develop their business with their core market in mind – in other words, they are very Southeast Asian in their approach. Gojek took the native Indonesian ojek or motorbike riders and digitized the service, making it vastly more accessible to drivers and customers alike. Regional e-commerce giants Shopee and Lazada won the market through its cash on delivery (COD) strategies, appealing to users who are either unbanked or are apprehensive about online payments.

At WIZ, we understand that Southeast Asian customers prefer to speak in their local languages, which is why we design our Talkbots to speak like the local consumer. We empower local businesses to embrace AI by tailoring our products to the market.

Why unicorns matter

Unicorns, being digital and growing by network effects, promote the expansion of internet access and the adoption of digital services in the region, which in turn drives the development of digital infrastructure. They also focus on addressing unique regional challenges and promoting financial inclusion, contributing to the development of a more vibrant and inclusive digital economy.

The focus on emerging markets is an important characteristic because by it, theses unicorns broaden the coverage of the startup ecosystem. Startups from other countries may target more mature markets like the US, Europe, or China, which have established consumer segments and industries. By working in developing markets, Southeast Asian unicorns can help shape these new markets and grow them significantly. They also focus on the emerging middle class that has an increasing purchasing power, which can determine the direction these markets take.

While the ecosystem is continually growing, I believe we need to encourage the growth of unicorns by creating conducive regulatory environments, developing digital infrastructure, and supporting a culture of entrepreneurship and innovation. We also need to strengthen regional collaboration and improve intellectual property protection.

Challenges for “soonicorns”

Aside from unicorns, there are “soonicorns” – startups that have the potential to become unicorns (valued at over $1 billion) but are not quite there yet. The rise of Southeast Asian soonicorns comes with as the region’s growing middle class, improving infrastructure, and supportive government policies. But there are also challenges which are somewhat embedded into the Southeast Asian market. These include market fragmentation, regulatory complexity, lack of access to funding, competition for talent and market share, and a lack of digital infrastructure. 

The lack of access to funding, particularly at the early stages, is something that can make it difficult for startups to get off the ground and compete with more established players. However, there are a number of initiatives aimed at addressing this issue, including government-backed funds and private sector investment.

Another challenge facing Southeast Asia’s startup ecosystem is the lack of talent. While there are plenty of young, tech-savvy graduates in the region, many of them lack the skills and experience that startups require. We experience this personally at WIZ. We need locals to do the work in the local languages, and we need them to have some experience in AI or at least have the capacity to grow in the space. And yet, despite this challenge, we have continued to grow and expand into markets that were practically AI deserts. We love that we can bring about this kind of change.

Southeast Asia’s tech ecosystem will likely see more unicorns emerge in the next few years, particularly in the field of AI. And with the rise of unicorns comes the inevitable growth in region’s economy.

ATxSG was jointly organised by IMDA Singapore and Informa Tech, held from 06-09 June 2023.

22
May
Talkbot Basics  ·  Voice AI Technology
The transformative role of Large Language Models in insurance
Jennifer Zhang

Jennifer Zhang

CEO & Co-Founder

In today’s rapidly evolving digital landscape, where technology continually shapes industries, Large Language Models (LLMs) have emerged as powerful tools that can revolutionize the insurance sector. Because a lot of the information around insurance – such as policies and claims – are textual, LLMs can play a big role in transforming the industry with artificial intelligence (AI).

Enhancing customer experience

In today’s customer-centric business landscape, providing exceptional customer experiences is paramount for insurance companies. LLMs that are trained on industry-specific language can play a crucial role in revolutionizing customer interactions and elevating the overall customer experience through AI technology.

AI solutions powered by insurance-specific LLMs can provide more personalized and efficient Customer Service.Through natural language processing capabilities, LLM-powered bots and virtual assistants can engage in conversational interactions with customers. LLMs enable these digital assistants to understand customer queries more accurately, and thereby provide more relevant and timely responses. At their best, LLM-powered virtual assistants can assist customers in filing claims and offering policy information, enhancing the speed and convenience of customer service.

Text-to-speech engines can also leverage LLMs to facilitate seamless multichannel communication. Instead of just interacting through text, customers can call and get their queries addressed by voice. Many customers prefer talking through critical conversations like requesting policy changes, and LLMs that feed into voice AI solutions like WIZ Talkbots can do just that. LLMs enable consistent and context-aware conversations across different touchpoints, ensuring a seamless experience regardless of the communication medium.

Then there’s localization. Companies that have their own LLMs can integrate local terms and accents to cater to their customers better, especially over voice channels.

Improving insurance processes

The insurance industry involves numerous complex and time-consuming processes that can benefit greatly from automation. Having industry-focused LLMs can help automate various aspects of insurance operations, offering increased efficiency, accuracy, and cost savings. Here are some use cases –

  • Underwriting and Risk Assessment:
    With LLMs, insurers can better assess risk profiles and make more accurate underwriting decisions based on a wealth of historical data.
  • Claims Management:
    AI can facilitate automated claims handling by extracting relevant information from claim forms, policy documents, and supporting evidence – all guided by an LLM that was designed and developed for insurance.
  • Settlement Negotiations:
    By understanding and interpreting natural language, insurance LLMs can generate settlement offers based on predefined criteria. This automation expedites the negotiation process, minimizes human bias, and enhances consistency in settlement outcomes.
  • Customer Onboarding and Policy Generation:
    LLMs play a crucial role in automating customer onboarding and policy generation processes. By leveraging natural language understanding, AI that’s powered by insurance LLMs can assist customers in providing the required information, auto-populate application forms, and generate policy documents with minimal intervention. 
  • Regulatory Compliance:
    LLMs aid in automating regulatory compliance processes by analyzing legal and regulatory documents, identifying relevant clauses, and ensuring adherence to compliance requirements. This helps insurance companies stay up to date with changing regulations, mitigates compliance risks, and minimizes manual effort in compliance monitoring. 

Jagadiri Insurance success story

One example of a successful deployment of voice AI solutions in insurance is from our client Jagadiri, a company under PT Central Asia Financial. With WIZ.AI Talkbots deployed for premium payment reminders, Jagadiri increased client contact rates by 40% and improved their cost efficiency 5.3 times. The solution is continually evolving and Jagadiri is now looking to use Talkbots for customer onboarding.

The example from Jagadiri Insurance shows how powerful localised, custom LLMs can be in AI deployment within the insurance industry. Looking ahead, the technology can provide further efficiency improvements and cost reductions, all while elevating the customer experience.

Explore how insurance-specific large language models can help you scale your operations. Speak to a specialist today.
Contact us

17
May
Talkbot Basics  ·  Voice AI Technology
A roadmap to successful, ROI-driven AI adoption for enterprises
Jennifer Zhang

Jennifer Zhang

CEO & Co-Founder

Artificial intelligence, specifically voice AI, has the potential to revolutionise the way companies interact with their customers. Companies that adopt voice AI solutions are poised to reap numerous benefits, from time and cost savings to increased customer satisfaction. But how do you figure out where to deploy AI solutions?

Finding the sweet spot for AI deployment

Some operations stand to gain more from voice AI or WIZ Talkbot solutions. Here are some guidelines to help you identify and select the optimal use case for your AI-powered solution. Since voice AI is most applicable to calls or voice interactions, we’ll focus on these areas.

  • Identify use cases that represent at least 10% of your total call volume
  • Find operations where interactions are highly transactional – these include one-off queries or simple troubleshooting or customer support interactions.
  • Zero-in on tasks that handle common tier-1 customer queries 

Once you’ve found a number of use cases for your Talkbot, it’s time to narrow down the selections. Identify one or a few use-cases with the highest potential return-on-investment (ROI) based on estimated savings. This will be the areas where you are likely to see more success in Talkbot deployment.

After identifying the first area of application, it may be worthwhile to identify parallel or adjacent functions where Talkbot adoption can be easily transferred. Once you are ready to scale and expand your AI application, you can adopt the existing solution to new use cases and apply the lessons learnt from the first adoption.

Set yourself up for success

Great outcomes start from laying a good foundation. And in voice AI deployment, there are some steps we recommend you take in order to make the most out of what the solution can offer – and generate results you can be proud of.

Acquire Data and Metrics – Look at any quantitative or qualitative contact center data you may have at your disposal. Example metrics are average handle time, cost per call minute, IVR resolution rate, etc. Pool and study these metrics to understand understand how you arrive at exceptional outcomes. You may also want to collect samples of calls from your top agents to benchmark performance.

Identify In-House Experts – Engage your in-house subject matter experts and stakeholders. These may include professionals who manage or write call scripts, call center managers, customer success managers. Seek their input and involve them in the adoption process early. Get them to help define success metrics, targets, and processes.

Define Success – Identify the key metrics and KPIs you currently track and how they map against the use case you’ve considered. Use insights from existing data to guide your strategy; take cues from top performers, key stakeholders, and subject-matter experts.

Map connected systems – Understand what other technologies and systems will need to be integrated or will be impacted by the voice AI solution. These may include your CRM, call center software, etc. Pull in your IT department early to get their thoughts on possible roadblocks or speedbumps. Ensure that IT is fully engaged and ready to support.

Obtain stakeholder buy-in

Success is a team effort. It’s always a good idea to engage with key stakeholders early in the adoption process to bring a depth to the conversation. Their support will be invaluable to making your shift to AI-powered customer engagements seamless. Some people you may want to speak with even while exploring AI solutions would be –

  • Executive Sponsor – Your executive sponsor is a critical stakeholder and can serve multiple purposes, including helping you navigate internal processes, influencing other decision makers within the organization, and functioning as a sounding board for ideas and strategy.
  • Customer Experience – Customer experience leaders take on different titles or functions in every organization, and it’s best to include them from the beginning to end. They’ll have the customer’s best interests in mind and provide critical feedback and context, especially in designing customer interactions.
  • Call Center Managers – Your call center and customer service managers often have the deepest level of contextual knowledge because of their experience on the field. It’s recommended to involve them from day one.
  • IT – Your technical stakeholders will be an important part of our implementation and deployment. The sooner you map out technical requirements, the better.
  • Procurement – Involve your procurement team as they’re often integral in vendor selection and registration. Procurement can help with the vendor registration process, tech risk assessment and negotiate commercial items in the contract.
  • Finance – Every project needs the backing of finance, from budget to savings. 
  • Legal – You need to make sure everything aligns with regulatory requirements and legal obligations.

Adopting an AI solution can be challenging, and you may have more questions than answers. By taking a more big-picture, proactive approach, you can ensure the success of your AI adoption and begin your organisation’s AI transformation journey. 

Looking to adopt AI as part of your business strategy? Our specialists can help you find the best solutions.
Contact us

12
May
Talkbot Basics  ·  Voice AI Technology
Game-changing communication: How LLMs are impacting Southeast Asian telcos
Jennifer Zhang

Jennifer Zhang

CEO & Co-Founder

The telecommunications industry in Southeast Asia is one of the fastest-growing and most dynamic markets in the world. With a population of over 650 million people and a rapidly expanding middle class, the region offers immense opportunities for businesses to tap into new markets and grow their customer base.

However, the industry is not without its challenges. The region is characterized by diverse cultures, languages, and economic conditions – all of which could hinder telcos from expanding their operations. Other challenges include inadequate infrastructure, limited access to financing, and regulatory hurdles. High competition in the industry puts pressure on companies to constantly innovate and improve their services to stay ahead of their competitors.

Despite these challenges, the telecommunications industry in Southeast Asia is poised for significant growth in the coming years. Home to one of the most tech-savvy demographics, the region is witnessing a significant shift towards digitalization. This trend has opened up new opportunities for businesses to leverage the power of domain-specific large language models (LLMs) to enhance their customer experience and gain a competitive edge.

Transforming telecommunications with large language models

Large language models (LLMs) are a type of artificial intelligence technology that can understand and generate human-like language. This capability opens up multiple possible use cases in the telecommunications industry, such as:

  • Improving customer service and support: LLMs can be used to analyze customer data, such as their chat histories and social media interactions, to understand their needs and preferences. This data can then be used to provide personalized recommendations and support, which can improve customer satisfaction and retention.

  • Enhancing chatbot interactions and voice assistants: LLMs can be used to improve the accuracy and naturalness of chatbots and voice assistants, making them more effective at understanding and responding to user requests. This can help to reduce the workload of human customer service agents and improve the overall customer experience.

  • Automated data analysis and insights: LLMs can be used to analyze large volumes of unstructured data, such as customer feedback and social media posts, to extract insights and trends. This can help telecommunications companies to identify areas for improvement and develop more effective marketing and sales strategies.

  • Increasing efficiency and productivity: LLMs can be used to automate repetitive tasks, such as data entry and content creation, which can free up human resources for more complex and strategic work. This can help to increase efficiency and productivity within the organization.

Forging the way ahead with localised enterprise LLMs

Because the Southeast Asian market is fragmented not just geographically but also linguistically, being able to operate across the region requires multilingual capabilities. This is one key area where LLMs are becoming increasingly proficient. WIZ.AI focuses on the localisation of artificial intelligence (AI) solutions like our flagship Talkbots and now our enterprise LLMs. The capacity for machines to communicate in multiple languages and local accents can help to break down barriers and improve communication with customers across different countries and cultures.

And with domain-based, region-specific knowledge, enterprise LLMs can have better contextual understanding. When businesses have their own LLMs, they are better positioned to take advantage of what generative AI has to offer. There is increased understanding of communications which enables more accurate and relevant AI-generated responses. Companies can leverage their own LLM not just for customer engagement – LLMs can be used for internal operations such as employee onboarding or drafting meeting summaries. Integrating LLMs with other technologies, such as chatbots, voice assistants, and natural language processing (NLP) tools, can provide more comprehensive and effective solutions that are closely linked to business operations.

Proving the point: Link Net success story

One recent example in the successful application of localised LLM is the use of WIZ Talkbots by PT Link Net Tbk. Link Net, am Indonesian telecommunications company that serves nearly three million households, improved its customer outreach by deploying Talkbots for its payment reminder and recontracting calls. Link Net was able to double its customer outreach and increase payment collections within one month of using Talkbots. WIZ.AI trained Link Net’s Talkbots with our localized Bahasa language model, proving that enterprise LLMs can greatly benefit large-scale telecommunications companies.

As AI technology continue to evolve and advance, businesses in the industry will need to stay up-to-date with the latest trends and developments to remain competitive and leverage the full potential of LLMs. By doing so, they can position themselves for success in an industry that is rapidly evolving and transforming.

Curious about how an LLM can be part of your telco operations? Speak to one of our specialists today.
Explore LLMs

09
May
Talkbot Basics  ·  Voice AI Technology
Unleashing the potential of LLMs: a new era for financial services
Jennifer Zhang

Jennifer Zhang

CEO & Co-Founder

Language Model Models (LLMs) are the latest buzzword of the tech industry, and for good reason. The most popular LLM today – OpenAI’s ChatGPT – has proven to be a powerful tool for day-to-day tasks, from summarising huge chunks of text to drafting emails. It’s no surprise that tech-savvy employees have started to use it to boost their efficiency at work. But using third-party general LLMs have risks and limitations. 

Recently, Bloomberg, the financial software, data, and media company, has launched its own LLM called BloombergGPT. The model is designed specifically for financial traders and investors, and marks a significant milestone in the use of LLMs in financial services. We’ll briefly discuss BloombergGPT’s capabilities and its implications on the financial industry moving forward.

Helping traders make sense of the news

BloombergGPT is specifically designed to help traders and investors analyze news and data, enabling them to make better investment decisions. BloombergGPT is able to detect whether news headlines indicate a bullish or bearish sentiment,  for example. The finance-specific LLM analyzes news stories, earnings reports, social media feeds, and other sources of information to identify relevant trends and patterns.

And similar to Bing Chat, BloombergGPT can answer questions like “Who is the CEO of Citigroup Inc?” – and do it fairly accurately. That’s because BloombergGPT is trained on 40 years of proprietary data that enables the LLM to identify corporations, business entities, and even stock ticker symbols. Bloomberg will be integrating its LLM into one of its core services, the Bloomberg Terminal, enabling subscribers to get even more value out of the search database.

BloombergGPT is a powerful example of how LLMs can be used to power the financial services industry, and its potential is only being discovered.

Revolutionising financial services

The example of BloombergGPT shows that LLMs have the potential to transform the financial services industry by automating and streamlining many processes. Finance-specific LLMs can analyze large amounts of data, identify patterns and trends, and make predictions with a high degree of accuracy.

Another way LLMs can be used in financial services is in fraud detection. Trained on historical data and typical transaction patterns,  LLMs can be used to identify potential fraud by detecting unusual activity. Rules can be set up for AI assistants to alert financial institutions when this occurs, enabling banks or investment houses to take immediate action like blocking transactions or accounts to prevent further damage.

A more common use case which is already being down with general purpose LLMs are AI bots in customer service. And while general LLMs already improve customer service efficiency, having a finance-specific LLM can improve outcomes even further. An industry-focused LLM can process specific finance-related customer inquiries and provide more fine-tuned responses. With finance-specific LLMs, AI bots may be able to handle more complex queries, further reducing the workload for customer service representatives and improving response times.

The future of banking and investment

There are various other ways to use custom LLMs to boost efficiency and streamline operations in finical services. It can be deployed for employee onboarding, providing essential information for new staff. In a highly regulated industry such as banking and insurance, domain-specific LLMs can be used to analyze regulatory documents and identify relevant sections that require action for compliance.

And while BloombergGPT is impressive, it only functions in one language – English. This limits its application into broader markets and geographies. The future of LLMs in financial services are multi-lingual, domain-specific models that perform well in narrow finance-related tasks and also general queries.

Finance LLMs are not new – in fact, a research paper on financial LLMs was already published six months before BloombergGPT was announced. Still, these domain-specific AI models have the potential to revolutionize the financial services sector. They can help automate and streamline  processes, improve accuracy and efficiency, and reduce the risk of human error. Financial institutions that embrace LLM technology will  likely gain a competitive advantage over their peers.

Want to stay ahead in the financial industry with your own LLM? Speak to one of our specialists today.
Explore LLMs

03
May
Talkbot Basics  ·  Voice AI Technology
The data security risks of using third-party generative AI tools
Jennifer Zhang

Jennifer Zhang

CEO & Co-Founder

Generative artificial intelligence (AI) tools like ChatGPT, Bard, and Bing search have revolutionized the way companies interact with customers, generate content, and perform other tasks. However, the use of these tools also raises significant concerns around data security and data protection. In this article, we’ll explore the risks of using third-party generative AI tools and offer some tips for ensuring that your company is using these tools safely and responsibly.

Generative AI and data security

One of the primary risks associated with using generative AI tools is the potential for data breaches or leaks. Because these tools are designed to generate content automatically, they often collect and store large amounts of data. This is the main reason why many financial institutions like JP Morgan Chase and Deutsche Bank have either restricted or outright banned the used of ChatGPT in their workplaces.

The risks involved in using these third-party generative AI aren’t obvious to many. Because ChatGPT has an iterative tool that relies on machine learning capabilities, any interaction with the bot is used as training data for future interactions. This means any information shared in the chat is saved in the AI engine’s servers – data which may include sensitive, personally identifiable information (PII). Because this data resides outside the user’s infrastructure, it will be hard to retrieve and delete if and when a data leak occurs.

The power of generative AI is impressive, which is why many people have embraced it and integrated it into their day-to-day tasks, such as drafting emails or generating code. However, this latter case was the cause for concern for Samsung, which has also banned the use of generative AI tools in the workplace. This comes after engineers have uploaded a source code into ChatGPT, potentially sharing sensitive intellectual property (IP) to a third party.

Using generative AI responsibly

The risks that come with using generative AI is largely tied to how people use it. Because these tools help increase our efficiency, it can be tempting to use them indiscriminately . However, once you understand how the technology works, you will start having some genuine concerns around data protection. Here are some steps to avoid data leaks when using generative AI tools –

  1. Limit the data you share: Only share the data necessary for the tool to function, and make sure that any sensitive data is either anonymised, made obscure (such as using percentages or using a factor of 10), or not shared at all.
  2. Train your employees: Make sure that all employees who use the tool are trained on data protection and privacy best practices. Help them understand the risks so they can avoid sharing sensitive IP or proprietary information.
  3. Keep data in-house with a local large language model (LLM): Instead of sending data to a third-party, you can use AI solutions that rely on local language models. This requires the development of an LLM specific to your company, done by an AI solutions provider.
 

Benefits of using a local large language model (LLM)

The third option to ensure the security and protection of company data is a path that was taken by Bloomberg, the financial services company. It developed its own language model called BloombergGPT, which is specifically tailored to the needs of the financial industry. The 50-billion parameter LLM has been built using Bloomberg’s financial data and domain-specific knowledge. Bloomberg’s AI team decided to leverage the rich data the company has amassed in 40 years, and translate that into training data for its own generative AI solution.

BloombergGPT is a great example of a domain-specific LLM that outperforms general-purpose LLMs in a narrow field – in this case, financial markets. And although Bloomberg may not have decided to develop its own GPT or LLM for security reasons, having it ensures that its employees won’t inadvertently share sensitive data outside the company. BloombergGPT is poised to help many of the company’s employees by coming up with accurate reports as well as process the vast amounts of data that come through the Bloomberg Terminal.

Generative AI tools offer tremendous benefits to companies looking to automate tasks and generate high-quality content quickly and efficiently. However, it’s important to recognize that these tools also come with risks around data security and data protection. Having your own company- or domain-specific LLM will help mitigate these data leaks while providing your employees the benefits of using generative AI.

Want to see how having your own business- or domain-specific language model can benefit your business? Contact one of our specialists today.
Explore LLMs

Page 1 of 5