It`s been more than a year since OpenAI launched ChatGPT, the chatbot everyone today is talking about. Despite AI is not something new, it`s been around for quite some time and many industries have already applied it successfully to boost their performance, productivity, and growth long before 2022, there is the feeling that the inauguration of ChatGPT has marked a whole new era of AI-powered innovations, digital products and applications that will not only help us in our everyday lives but transform, perhaps even disrupt whole business sectors and industries. But if AI has long been used in businesses and products, what makes the current time different? Throughout the years, there have been many deep-learning algorithms that were developed and trained in a way that they could learn from big sets of data, spot patterns, and automate and optimize business processes. Now, we have a pre-trained natural language model that can learn from vast amounts of text and data, and not only spot patterns, but generate human-like responses and understand human speech. What is more, it has never been as accessible to the general public as it is now (its API is open for developers to take advantage of the model for their products).
That`s enough to create a buzz in the space and spur a new wave of innovations. A proof of how big the hype is and its potential to cause disruptions within industries is the fact that it has only been recently since governments around the world have passed their first AI legislations while debates on its regulations are actively being carried on. Will these new regulations cause a slowdown in this industry? It could be, as they`re putting an additional burden on start-ups developing new AI models who will soon have to comply with certain sets of rules and criteria, driving their costs up. But it`s either companies will become more creative in their approaches and how to fit into these new criteria, or this will lead to asserting the market positions of these few really big models and laying the foundations of competitive barriers for smaller entrants. Even in the later case (which is more probable), I guess that start-ups can continue coming up with new, innovative AI-powered digital products simply by using the ChatGPT or Bard APIs. Of course, we expect to see how this situation will roll out in the future. Now, let`s dive into the momentum and see how AI is innovating and transforming digital products for the end users at the moment.
I reviewed four different industries/markets to get you a snapshot of what is currently out there These are: personal development, healthcare, fintech, and martech.
1. Personal Development
The market for personal development is a bit non-standard, but аn intriguing and fast-growing industry where one can see some very interesting AI-powered developments. According to Grand View Research, for the year 2022 it was estimated to have been $43.77 billion, while at the same time, it is anticipated to grow at a CAGR of somewhere around 5.5% between the period of 2023 and 2030 (on a global level). But what exactly the personal development market is and what does it include, to many, it may sound like a vague and wide concept. In their calculations, the researchers have included the following domains:
- Skill Enhancement
- Physical Health
- Mental Health
- Motivation & Inspiration
- Self-awareness
Whereas the product types (or instruments) are split into the following categories:
- Personal Coaching/Training
- e-Platforms
- Workshops
- Books
Some of the key drivers of the personal development market growth are individuals` increased focus on self-improvement through enhancement of their professional and social skills, better self-awareness, the development of behavioral and personal traits that align with their goals and aspirations, as well as the emphasis they put on their physical and mental health. In a nutshell, their desire to be а “better version” of themselves. The latest breakthroughs in AI and conversational language models have allowed whole new opportunities in this area, the most common of which being the virtual therapists, trainers, and coaches one can get at a no-brainer cost in comparison to what a real human professional would charge you. What is more, you can have sessions anytime 24/7, whenever you feel like it, or whenever it is most convenient for you. And while I have to admit some newly developed chatbots are not as good as their developers intended to make them, or as a user would wish (don`t forget there are real-human therapists/coaches/tutors that are not-so-good either), others exceeded my expectations in their capability to comprehend various in-depth issues and problems and suggest solutions or concrete action plans (with some guidance on my part, of course). After I tested dozens of applications, If i have to sum up my observations on their performance, I would say that some could be a viable alternative to a real human professional, with the main benefits being their cost-efficiency, convenience, capability to converse as if you are talking to a real person and to create a data visualization of insights they extracted from your interactions. However, I still believe (especially when it comes to therapists), that real humans could do better than their respective chatbots in many cases at the current point of time. Unless you guide them, chatbots are not always capable of understanding certain situation-based contexts. For example, life in different geographical areas varies and can have its peculiarities that virtual bots do not understand and they sometimes provide you with a response you cannot relate to. We can suppose that the more information they accumulate over time and the more they learn, the more polished they will get, but we don’t know for sure how far they could go. Will they be able to outperform real human professionals? That`s a million-dollar question with proponents of both sides of the theory.
Here are some examples of innovative, AI-powered apps and products part of the personal development market that I really liked and enjoyed and that I believe, work well.
Deepen is an AI companion and therapist that cost less than 15euros a month (with a 7-day free trial period). It is a mobile app, available to download on App Store and Google play. What I liked about this app is the fact that, unlike some other AI therapy and counselling apps, you can have as many sessions as you need or want, and it “remembers” your previous conversations. That means, with each interaction or session, it gets to know you better and it suggest a more personalized response or solution.
Once you intall it, you`ve got few onboarding questions (about how you feel, what communication style you prefer, what are your goals, etc), along with a brief intro to the capabilities of the app. After you are all set up, which should not take more than 2 minutes, you can start a chat with your virtual companion (there is an option for voice messaging if you prefer speaking), just like you are chatting with a real human therapist. The chat itself is at the core of the app`s functionality. The other (secondary) features include a visual board of metrics on your mood, energy and stress levels over the time, and an insights dashboard highlighting the main themes and mental states drawn from each session. In other words, the AI machine is capable of measuring your mood by the way you are conversing and identify and synthesize those themes that are central to you.
If i have to sum up my experience, the chat bot is better than I expected. I tested few hypothetical examples, and it provided me with some quite good mental and stress relief techniques tailored to the specific situation, as well as solutions to a few more complex issues I made up. It also proved to be capable of preparing quite detailed, step-by-step plans on how to achieve certain goals in my life by taking into account specifics and limitations related to the situation, of course with some additional guidance on my side.
Similarly to Deepen, Youper puts its chat bot at the core of its app. I tested it with scenarios close to the ones I used for Deepen and the bot is just as good, in some cases even better. This may come down to the fact that they base their therapy approach on clinically-proven behavioural therapies, such as CBT, ACT, PST, etc. They have a proprietary technology that customizes and augments large language models to create a specialized application that guides users through their emotional challenges and allows it to adapt its approach to the above-mentioned behavioural therapies.
Once installed, there is again a personalized mental health assessment as part of your onboarding process. The test is more exhaustive and comprises greater number of questions, which makes it a bit more time-consuming, but it allows Youper to get better personalized to your needs and monitor for your specific conditions and mental challenges. After you complete the initial onboarding questions, you can start chatting with the bot. You are encouraged to check-in to the chat on a regular basis/ at a time interval you`ve picked, and even if you don`t feel like chatting or you don`t have the time to do so, you can have a quick check-in every now and then, where you provide a brief information on what is your current mood or the challenge you are facing. These brief, regular check-ins help track mood metrics better and build more accurate mental health assessment. The main drawback I identified is that, while the app attributes any mood information you provided during your conversations to the metrics and insights tab , it “filters out” any other personal details you shared once the given session has ended. In other words, unlike Deepen, it does not “remember” your previous conversations and the information you provide today will be lost in your next session, so for many things, you`ll have to explain yourself over and over again (which puts it in a disadvantaged position when compared to a real-human therapist). Another pitfall is that, despite you can have limitless number of chat sessions, it`s not always you who decides when the “conversation” ends. Your chat can get disconnected at any point should the bot decide you received enough support/ideas/information or for some other, unknown reason, which puts certain limits to the depth of conversation you can have, making it again more unsophisticated than a real-human professional.
In addition to the chat bot, there is a separate functionality that tracks six different mental health conditions. You can get tested for the following: Anxiety, Borderline Personality, Depression, Panic, PTSD, Social Anxiety. Upon completion of any of these tests, you`re assigned with a score that measures your symptoms from mild to very severe and you are encouraged to retake the test in a given time internal, which tracks your progress for the specific conditions.
As for the insights and reporting tab, you`ve got a simple diagram indicating what your mood has been during the week, and insights cards on what makes you feel in a certain way that get unlocked the more information is accumulated during your chats. Its aim is to help you build awareness of yourself.
Clarmind is a highly personalized AI therapy app, which is something fresh and different from the standard AI therapy apps you can find on the market, the majority of which are chatbots. In reality, Clearmind is not a chatbot in the sense I understand it and they do not put it at the heart of its app unlike their other competitors (as far as I can see, they currently don`t have a chat functionality on Android, but will be rolling it out soon).
So, how does it work? You have a text box, where you can type in what bothers you, express your thoughts and feelings and describe your situation. The more details you input, the greater the output will be. Once you`re done and you`ve clicked “Try Clear”, you`ll get information and guidance personalized to you and your case in a few different formats:
– Moodcard: this is a visual representation of your feelings and situation. It assesses the information you provide, draws conclusions from it, and organizes it into the following categories to make it easier for you to understand your underlying issues and get better self-awareness:
Mood: what your current mood is. For ex. Overwhelmed, Unhappy, etc.
Trigger: What triggers you to feel that way
Focus: based on your input, what you are focusin on in your life right now. The main theme
Personality: what your personality is, traits that are central to you
Mental Profile: your mental health state
Environment: what is your environment? This is related to the people around you. For example, whether your environment is supportive or support-lacking
Habit: at first this sounds a bit unclear and tricky, but I understand it the following way – based on the info provided, what habit should you employ in your daily life to empower yourself.
Along with the information above, you get a thematic image, which illustrates your case and includes a few keywords that best describe you.
– Response: Unlike the standard AI chatbot therapists, you get a personalized response in the form of a long paragraph. You can`t answer it back. This paragraph aims to help you and suggest solutions to your problems. It analyzes your issues and mental state and proposes recommendations on how to act and what to do to improve your life and mental health.
– To-do: While the response paragraph you get creates kind of a plan on what to do in order to solve your problems and feel better, the “Your to-do” feature automatically finds you perfect goal and breaks it down to specific tasks you can undertake to lay down the groundwork towards completing your goal and achieving success. In my case, it looks like a step-by-step plan, where each task is designed to be doable within a day. In my opinion, this is one of the coolest features in their functionality stack.
– Meditation: The app can generate a personalised script you can use to meditate.
– Hyper-personalized recommendations: This is not part of the main tabs at this point, but is part of the app. These are suggestions, in the form of songs, podcasts, TV shows, and movies to watch or listen to, that are aimed to make you feel better, motivate and inspire you. There is a rationale included behind each suggestion here that justifies it.
To sum it up, to me, Clearmind is more than just a mental health app. It is a tool that guides and inspires you towards achieving your wants and goals. Moreover, its design and features are something we haven`t seen in any other app, which makes them innovative and full of potential for growth!
The above three examples are part of Mental Health mostly, and then Self-Awareness and Motivation & Inspiration niche markets, but as we already mentioned, the Personal Development industry is much more than that. Therefore, let`s look at another type of product, which is also representative of this wider market category.
Whoop is a fitness-tracking wearable and app that syncs with it, and is one of the most famous in this niche. The device tracks and gathers biometric data, such as heart rate, skin temperature, respiratory rate and blood oxygen (they`re constantly looking for new ways to enhance their tracking capabilities), which is then used to learn about your body`s patterns and baselines. Based on this knowledge and the personal goals you`ve set for yourself, Whoop, by using AI, provides you with insights and personalized recommendations, such as how much sleep and rest you need to recover, as well as how to tailor your lifestyle, training, and habits to unlock your fullest potential.
Apart from this, in September 2023, Whoop launched its new AI conversational feature, powered by the GPT-4 technology – Whoop Coach. Whoop Coach is a chatbot that provides personalized human-like answers to your health, fitness, and wellness questions based on the millions of data points that Whoop already knows about you and the Chat GPT`s natural language processing model capabilities. You can access Whoop Coach by heading to your Home screen or from any of the key pillars in the app like Sleep, Strain, and Recovery. You can ask it anything you want and it will respond, evaluating your personal Whoop data, that of the Whoop community, and the latest performance science and research.
A good example of how AI is transforming the skill enhancement market niche is through the various language-learning chatbots that have been developed recently which can help you improve and even master a foreign language. People who want to improve their foreign language skills are now given the unique opportunity to practice without having to look for another speaker of the same language and most importantly, at any time.
One such app is LangAI, powered by ChatGPT and currently available on the App Store only. It allows you to chat and speak in over 20 different languages, among which are Italian, Spanish, German, French, English and more. LangAI provides live corrections and instant feedback during your conversations on your pronunciation, grammar, and vocabulary. Conversations are adaptive and personalized to rapidly enhance your language skills and fit into your learning style, goals, and objectives. LangAI helps you practice speaking, listening, reading, and writing in your target language. Sounds cool, isn`t it?
2. Healthcare Industry
One of the other domains where AI is having a significant impact is healthcare, where it can help improve diagnosis, treatment, prevention, and management of various health conditions.
Here is an example of some of the recent applications of AI in the healthcare industry:
– Medical imaging: AI can help analyze images such as X-rays, CT scans, MRI scans, and ultrasound to detect abnormalities, measure anatomical features, and assist in diagnosis and prognosis. For example, AI can help diagnose breast cancer from mammograms, detect lung nodules from chest X-rays, and segment brain tumors from MRI scans.
– Drug discovery: AI can help accelerate the process of discovering new drugs by screening large databases of chemical compounds, predicting their interactions with biological targets, and optimizing their properties. For example, AI can help identify potential candidates for treating COVID-19, Alzheimer’s disease, and cancer.
– Clinical decision support: AI can help clinicians make better decisions by providing them with relevant information, recommendations, and alerts based on the patient’s data and medical literature. For example, AI can help predict the risk of complications, infections, or readmissions for patients undergoing surgery, suggest the best treatment options for patients with chronic diseases, and alert clinicians about potential adverse drug events or drug interactions.
– Personalized medicine: AI can help tailor the treatment and care to each patient by taking into account their genetic makeup, lifestyle, preferences, and response to previous interventions. For example, AI can help recommend the optimal dose of a drug for a patient based on their pharmacogenomics profile, suggest the best time to administer a vaccine based on their immune system status, and adjust the parameters of a pacemaker based on their heart rhythm.
However, even though AI is transforming the healthcare industry by offering great new possibilities and opportunities for improving health outcomes, and drug and disease research, as well as the organization of various healthcare institutions, we have to note that it also poses certain challenges and risks that need to be addressed and carefully examined. Such are:
– Data quality and privacy: AI relies on large amounts of data to learn and perform its tasks. However, the data may be incomplete, inaccurate, biased, or outdated. Moreover, the data may contain sensitive personal information that needs to be protected from unauthorized access or misuse.
– Ethical and legal issues: AI may raise ethical and legal questions regarding its accountability, transparency, explainability, fairness, and human oversight. For example, who is responsible if an AI system makes a wrong diagnosis or prescribes a harmful drug? How can an AI system explain its reasoning and justify its decisions? How can an AI system avoid discriminating against certain groups of patients or violating their autonomy?
Now, let`s delve into some of the companies that are examples of developing and offering incredible AI-powered products for the healthcare industry.
1. IBM Watson Health
IBM Watson Health is a division of IBM that uses artificial intelligence (AI) to provide solutions for the healthcare industry. IBM Watson Health aims to improve patient care and operational efficiency by harnessing the power of data, analytics, and automation. Some of the ways that IBM Watson Health works and uses AI are:
- Imaging AI Orchestrator:
This is a cloud-based service that offers imaging organizations access to regulatory-cleared AI applications from various providers. These applications can help radiologists with tasks such as detecting anomalies, measuring lesions, and prioritizing cases. The AI Orchestrator integrates the AI applications with the existing picture archiving and communication system (PACS) and delivers the results to the radiologists within their workflow. - Imaging Workflow Orchestrator with Watson:
This is a solution that modernizes the radiologist’s reading experience by providing a unified worklist that aggregates studies from multiple sources, such as PACS, vendor-neutral archives (VNA), and cloud storage. The solution also leverages AI to optimize the worklist based on factors such as urgency, complexity, and subspecialty. - Watson for Oncology:
This is a cognitive computing system that analyzes large volumes of medical literature, clinical guidelines, and patient data to provide evidence-based treatment recommendations for cancer patients. The system also provides supporting information such as confidence scores, relevant studies, and potential side effects. - Watson for Genomics:
This is a system that analyzes genomic data from tumor samples and compares it with medical literature, clinical trials, and guidelines to identify potential targeted therapies and clinical trials for cancer patients. The system can process complex genomic data in minutes, which would otherwise take weeks for human experts. - Watson Care Manager:
This is a cloud-based platform that helps care managers coordinate and personalize care plans for individuals with chronic or complex conditions. The platform uses AI to analyze data from various sources, such as electronic health records (EHR), claims, social determinants of health, and patient-reported outcomes. The platform then provides insights and recommendations to help care managers optimize care delivery, engage patients, and reduce costs.
2. Google Health
Google Health is a division of Google that focuses on applying artificial intelligence to healthcare research and technology. Some of the products that Google Health has developed using AI are:
- ARDA:
ARDA, which stands for Automated Retinal Disease Assessment, is an AI system that can help healthcare workers detect diabetic retinopathy, a leading cause of blindness, from an eye scan. It uses a deep neural network that was trained on more than 100,000 retinal scans that were manually reviewed by a large team of ophthalmologists, and as a result, is capable of detecting signs of diabetic retinopathy from new scans.
- Mammography
This is another AI product developed by Google for their Google Health stack. Its main goal is to help ragiologists detect breast cancer more accurately, quickly and consistently. accurately, quickly, and consistently. The system is trained on thousands of de-identified mammograms from female patients from the UK and US using Google’s AI technology to learn the complex features in mammograms that are likely to represent signs of cancer. As a result, it is claimed that it may spot signs of cancer that some specialists might not be able to see.
- DeepVariant
An open-source software that uses a deep neural network to identify genetic variants from DNA sequencing data. It`s been developed as a result of few years-long research collaboration between the Google Brain Team and Verily Life Sciences.
3. BenevolentAI
BenevolentAI uses AI to help discover and develop new medicines for various diseases, such as Parkinson’s, Alzheimer’s, and COVID-19⁴. It uses AI to mine and integrate vast amounts of biomedical data, such as scientific literature, clinical trials, and molecular structures, and generate novel hypotheses and insights. It also uses AI to design and optimize potential drug candidates, predict their efficacy and safety, and accelerate their clinical development.
4. Nuance
Nuance is a company that uses AI to enhance the productivity and quality of clinical documentation and communication. It offers various solutions, such as Dragon Medical One, which is a cloud-based speech recognition software that allows clinicians to create and edit medical notes using their voice; PowerScribe One, which is a radiology reporting platform that uses AI to generate accurate and structured reports from medical images; and CDE One, which is a clinical documentation improvement tool that uses AI to identify and resolve documentation gaps and errors.
3. MarТech
The Martech industry is another area where Generative and Predictive AI plays a significant role in revolutionizing products and tools, as well as helping companies, marketers and brands automate and optimize their processes, campaigns and analytics, resulting in higher ROI.
So, what does Martech refer to, and what kind of tools and software it comprises of? Here is a brief description from marketingevolution.com that is pretty straightforward and easy to understand:
“Marketing technology, also known as MarTech, describes a range of software and tools that assist in achieving marketing goals or objectives. When a marketing team utilizes a grouping of marketing technologies, this is known as their marketing technology stack.”
So,basically these are any tools you use to market your product or service and facilitate your growth.
Here is a glimpse of how AI has transformed Martech tools and what new capabilities it added on. Basically, we can split them into four main categories:
1. Predictive analytics: AI-powered predictive analytics tools help marketers analyze vast amounts of data, predict customer behavior, make data-driven decisions, and personalize marketing messages, resulting in higher engagement rates and increased ROI. This includes marketing retention as well.
2. Ad targeting: AI is improving the efficiency and accuracy of ad targeting, reducing wasteful spending, and improving ad relevance for consumers
3. Automated tasks: AI-powered tools automate mundane tasks and free up time for marketers to focus on more creative and strategic tasks.
4. Chatbots: AI-based chatbots help marketers provide personalized customer service and build customer loyalty. They are similar to advanced sales reps who understand customer needs and guide them to the best option.
Some companies and tools that have taken advantage of the new AI technologies and have already successfully adopted them either within their existing functionality stack or by creating a new products entirely based on AI, include:
ChannelMix is a tool for marketing analytics that employs an AI technology to predict and optimize the ROI for various media channels. It uses a data automation platform that collects, unifies and processes marketing data from multiple channels and campaigns, all in one place. Its advanced ML algorithms can ingest, clean and transform data into powerful insights that will show you a holistic view of your customer journey with each touchpoint, as well as the successfulness and efficacy of each channel you are running campaigns on to inform you which ones have the biggest impact on realizing conversions and what share of impact each channel has. In that sense, by using its advanced algorithms for data unification and analysis, ChannelMix provides a powerful attribution model (they call it “Marketing Impact Modelling”) that they claim is capable of addressing the shortcomings of the traditional attribution models (such as the first- or last-click attribution, linear, time-decay, etc) by illustrating more accurate picture of what each channel real attribution is. Moreover, unlike traditional models, they incorporate online and offline sources, such as phone calls, in-store visits and point-of-sale transactions for added precision and they`re using cookieless tracking technology to capture online data in a more reliable way.
However, its true AI Generative capabilities lie down to their channel mix calculator. By using the data the system already has about your marketing channels and campaigns or users behaviour and their purchasing history, not only you are given valuable suggestions on which channels you should focus on and direct your ad spend to, but you can also tweak certain variables (such as your ad budget or target revenue) to get the most optimal channel mix for the variables you predefined. For example, say you want to know how you should split your ad spending across different channels if you had X amount of budget and you want to achieve an ROI.
The main drawback of ChannelMix as I see it, is the fact that although it uses state-of-art AI technology to generate valuable insights and suggestions, it does not provide a level of automation that we`d expect to see, mainly because of their cookieless way of tracking. Their tracking involves manually appending a unique identifier (called a Key), to the end of any field in each campaign set-up you are creating. This Key serves as link between the campaign data and the metadata on this given campaign that is stored in ChannelMix Keys. You then need to once again manually create individual records on each campaign on your ChannelMix account and insert all the campaign`s details and parameters by yourself, so that it can start collecting data and analyze it. For those of you who expect a functionality where the tool can automatically adjust your mix of channel and campaign spending in live, for good or for bad, this is not the case with ChannelMix. It just provides insights and suggestions.
Intellimize is an AI-driven tool that helps you optimize and personalize your website for different types of visitors and users, enabling you to convert your traffic into results. It uses ML to automatically test and show the best website variations to each visitor based on their behavior, context, goals or certain rule-based criteria. Their proprietary AI is also trained on over 10 billion web pages making it a powerful optimization and personalization engine.
One of their best features is the so-called AI auto-discovery of segments. It can help you find and target the most relevant and profitable segments of your website visitors. It uses machine learning to automatically analyze your traffic and identify the patterns, behaviors, and attributes that correlate with your goals. It also shows you the performance and potential of each segment, as well as the best website variations for them. The cool thing is you don’t have to analyze your traffic data by yourself and manually define or guess the segments – the tool does it for you.
Another cool AI-based feature is their AI content generation tool called AI Content Studio, which can quickly create ready-to-use content for all conversion touchpoints on and off your website. AI Content Studio can generate content across the following categories:
- Website copy: This includes headlines, subheadings, calls to action, and other elements that persuade visitors to take action on your website.
- SEO-friendly product descriptions: This includes descriptions that highlight the features, benefits, and value propositions of your products or services, while also using relevant keywords and phrases to rank higher on search engines.
- Tailored landing page content: This includes content that matches the intent, context, and goals of your visitors, based on where they come from, such as search engines, social media, email campaigns, etc.
- Fresh blog topics: This includes topics that are relevant, interesting, and engaging for your target audience, based on your niche, industry, or keywords.
- Blog post content: This includes content that covers the blog topic in-depth, provides useful information, and encourages readers to take action or share your content.
- Social media posts: This includes posts that capture the attention, interest, and emotion of your followers, while also promoting your brand, products, or services.
- Ad copy: This includes copy that entices potential customers to click on your ads, while also conveying your unique selling proposition and value proposition.
AI Content Studio helps creating better converting content, by using generative AI that learns from your website data and goals.
Dstillery is a tool that provides custom audience solutions for brands and agencies. They use AI and data science to find and target the best prospects for programmatic advertising campaigns through their patented id-free technology, which means there are no cookies or third-party data used, allowing advertisers go ahead of any privacy concerns and emerging privacy regulations, and, at the same time, reap the benefits AI offers in ad targeting. Sounds innovative, doesn`t it?
Basically, by using their platform, brands can combine their own first-party data with an audience (or audiences) extracted from Dstillery`s platform ID-free® Custom AI targeting and use it on a DSP for programmatic campaigns.
What their ID-free technology actually is and how does it work? Generally, what it does is that It predicts the value of an impression to a brand without knowing anything about the user. This is possible through the following two mechanisms they have adopted:
- AI that is learning from browsing patterns detected in de-identified opt-in panel data. This means that they use data from a group of online users who have agreed to share their anonymous web activity with them for their AI to be trained on it.
- A neural network they have created, called MOTI (short for “Map of the internet”) can see how different sites over the net are linked and related to each other.
When they target users on internet, they do not have their personal information and they do not know anything about the users, but they use other signals like their geographical area, the URL they`re visiting and the time of their visit. These real-time signals are then combined with the above mentioned user browsing data on which their super computer was trained on, as well as their map of the internet (MOTI), to identify certain patterns and figure out user`s intent what he/she may be interested in or looking to buy. That`s how relevant ads are served to the right audiences.
Mnemonic AI is a tool that uses AI Generative technology to come up with some very interesting assets, such as user personas, for your brand, that may otherwise take you weeks of research survey, focus groups, interviews research to figure out.
It uses large language models and deep learning to analyze your own proprietary data, such as web analytics, email, CRM, call-logs or other, and can also enhance it with additional data sources that are publicly available, to generate the desired output for you, which takes around 72 hours, depending on the data volume. This output can be any of the following:
1. AI-generated Personas
AI-generated Personas help you understand the personas of your customers and products, and get deep insights into what drives them to interact with you and understand their needs.
2. Psychographic Segmentation
Psychographic Segmentation is a powerful tool to get meaningful insights into what your customers value and believe
3. Customer Personality
Customer Personality helps you understand how your brand personality is aligned with your customers and how they perceive you
4. Digital Twin of the Customer
Digital Twin of the Customer is a fully immersive representation of your customers, ready to answer all your questions in real-time, impersonating their personality, wants, and needs
CallMiner is a software that provides conversation intelligence solutions to businesses and call-centres. It provides a platform that analyzes omnichannel customer interactions at scale, using artificial intelligence and machine learning. Their technology transcribes conversations from various sources (such as voice calls, chats, emails, SM, etc), unifies the data gathered and analyze it to come up with deeper customer intelligence and personalized recommendations, in which way it improves the overall business performance of the companies.
The main solution through which CallMiner leverages the power of AI is their Eureka Platform. It is the tool where all customer conversations are captures, analyzed and that provides insights into customer sentiment, emotion, effort, intent, and satisfaction, as well as agent performance, compliance, and quality. In addition to the Eureka Analyze suite that show this key customer information, it offers the following AI-powered suites:
– Eureka Coach: A web-based application that enables managers and agents to review, score, and provide feedback on customer interactions. Eureka Coach helps improve agent engagement, retention, and performance, as well as customer satisfaction and loyalty.
– Eureka Alert: A real-time monitoring and alerting tool that identifies and notifies agents and supervisors of critical events or issues during customer interactions. Eureka Alert helps reduce customer churn, increase retention, and enhance service quality.
– Eureka Visualize: A data visualization and storytelling tool that transforms conversation analytics into interactive dashboards and reports. Eureka Visualize helps communicate insights and recommendations to stakeholders and drive action across the organization.
Adpost’s AI chatbot solution is affordable, yet robust. The chatbot can be customized and trained using specific information from webpages, FAQ lists, and Adpost’s ads. It also has the capability to handle multiple languages, expanding businesses’ global reach and enabling them to engage with customers from diverse linguistic backgrounds. The chatbot identifies situations where human assistance is necessary, such as complex queries or personalized advice. In these instances, the chatbot seamlessly transfers the conversation to a human representative. The platform ensures smooth onboarding, whether you’re new to chatbots or upgrading from a different system.
This is another cool SEO tool powered by the new Generative AI technologies that has been launched in November. AI Generate uses large language model technology to enhance content and create SEO-ready, on-brand content with just a few clicks. It analyzes content and provides actionable improvements to boost SEO performance by integrating high-impact keywords into the copy automatically. AI Generate also helps build a consistent brand voice and quickly apply it across the entire content experience.
4. FinTech
Another popular area where AI has taken over a significant part is the FinTech industry.
FinTech can be defined as any software that uses digital technology to provide financial services or solutions. Examples of fintech products include:
- Payment apps: These are apps that allow users to send or receive money digitally and often using mobile devices, such as PayPal, Revolut, Monese. They can offer different type of payment account and are alternative to traditional banks
- Peer-to-peer (P2P) lending apps: These are apps that connect borrowers and lenders directly, without intermediaries such as banks
- Investment apps: These are apps that allow users to buy, sell, or trade various financial assets, such as stocks, bonds, cryptocurrencies, etc.
- Crypto apps: These are apps that enable users to access, store, or transact with cryptocurrencies, such as Bitcoin, Ethereum, etc.
- Budgeting apps: These are apps that help users track their income, expenses, and savings, and provide insights and recommendations to improve their financial habits.
Here are just few cases in which the Fintech industry is leveraging AI technologies to transform itself, along with some exemplary software products and tools:
1. Credit worthiness analysis
These are applications or software that use AI to analyze various financial data sources and offer customized loan products and interest rates based on the risk profile of the clients. This use case of AI allows lenders to change the way they assess and manage credit risk. Some of the exemplary tools include:
This is software that uses AI techniques derived from genomics and particle physics to provide lenders with non-linear, dynamic models of credit risk that radically outperform traditional methods.
This is a software that uses AI to automate credit reviews, integrate with credit agencies, and predict and manage blocked orders. It also provides online credit applications, public and private financials extraction, and AI-driven credit risk monitoring.
This is a software that uses AI to build and deploy highly accurate credit scoring models that can handle complex and large-scale data. It also helps lenders explain their credit decisions and comply with regulations.
This is a software that uses AI to automate the entire lending process, from origination to servicing. It also offers credit scoring, fraud detection, loan management, and reporting features³.
2. Fraud detection and prevention
AI can detect and prevent fraudulent transactions and activities by using advanced algorithms and machine learning models that learn from historical and real-time data.
Below are few applications that use AI to automatically monitor transactions and events in real time to detect and prevent fraudulent activities occurring in-house, online or in-store.
This is a fraud detection tool that uses AI and machine learning algorithms to identify and prevent fraudulent activities. It offers features such as device fingerprinting, behavioral analysis, and anomaly detection.
This is a chargeback management tool that uses AI to automate the process of disputing and preventing chargebacks. It also provides real-time analytics and reporting features.
This is a credit-to-cash solution that uses AI to improve cash flow, operational efficiency, and risk management. It also provides Dun & Bradstreet data, analytics, and technology to help with credit intelligence and collections management.
This is a data analytics platform that uses AI to enhance banking fraud detection by helping their data analytics software recognize potential fraud cases while avoiding acceptable deviations from the norm.
3. Chatbots and robo-advisors
AI can provide automated and interactive customer support and financial advice through chatbots and robo-advisors that use natural language processing and understanding to communicate with users and offer relevant solutions.
Finchat.io is a fintech product that offers a generative AI tool for investment research. It allows users to interact with financial data, transcripts, filings, lessons from successful investors, and company-specific information in a conversational way.
Finchat.io leverages AI for its chatbot and robo-advisor by using natural language processing and understanding to generate accurate and easy-to-understand answers to complex questions. The chatbot uses OpenAI’s GPT-3 model to create text based on user input, and the robo-advisor uses Stratosphere’s proprietary algorithms to provide insights and recommendations based on market trends and patterns.
Betterment is a fintech product that offers automated investing, high-yield cash accounts, and retirement planning. However, they use AI selectively as they have experimented and found that neural networks for portfolio optimization for example, results in it lower expected results when compared to some simpler and statistical techniques. Nevertheless, Betterment uses AI to provide some personalized recommendations and insights for its customers, such as how much to save, how to allocate certain assets, and how to plan for retirement.
Robinhood also uses AI selectively and only for some of its features and functionalities. For example, their robo-advisor is AI-powered to provide its customers with personalized recommendations and insights. They tailor their suggestions and information based on customer`s individual preferences, goals, and risk profiles. It can recommend or ETFs that match customer`s interests or show how their portfolio compares to others
4. Portfolio management and trading
AI can help investors and traders optimize their portfolio performance and execute trades by using predictive analytics and deep learning to generate insights and recommendations based on market trends and patterns.
This is an advanced portfolio tracker that offers AI-driven portfolio management, market insights, and personalized recommendations. The platform aggregates all assets in one place, provides metrics to measure portfolio strength, and finds risks and problem areas to improve with automated recommendations.
- MT5:
This is a trading platform that supports third-party trading software that connects to dozens of online brokers. Its MQL5 marketplace lists thousands of AI-backed robots that can trade 24/7 on your behalf. You’ll need to purchase a robot and upload it into the MT5 software. There are AI robots for forex, stocks, gold, cryptocurrencies, and other popular markets.
This is a trading platform that allows you to leverage artificial intelligence to automate trading strategies. It’s user-friendly and requires no coding knowledge – you simply need to provide the AI bot with text prompts. For instance, you can instruct the bot to buy $500 worth of Amazon stock when the RSI hits 75, alongside your preferred stop-loss and take-profit.
This is a new AI trading tool that helps consumers make smart investment decisions. It can build a portfolio of stocks and funds, suggest suitable strategies to meet investment goals, and help you reduce your risk exposure. Magnifi comes as an app for iOS and Android and connects to several online brokers, such as Robinhood, E*TRADE, and TD Ameritrade.
About the Author:
Desy Paunova
Support My Work and Help Me Keep Creating!
If you enjoy the insights and analysis I share on product management, consider supporting my work. Your contributions help me continue to bring you in-depth content and keep the blog going strong. Every donation, big or small, makes a difference. Click the link below to support me and help shape future articles. Thank you for being part of the journey!
