The Rise of Digital Humans
Synthetic Media driven bots are revolutionizing online businesses
A few years back, it was difficult to imagine a world in which we can interact with digital humans on a daily basis. But as of today, you can encounter a digital human as a part of a layered digital fabric integrated into a website’s customer support bot or in a game you are playing. You will only realize the intensity of the artificial human interaction if you are unable to distinguish the bot from a real human.
A digital human can be of three types:
- A text-based chatbot that mimics a human conversation from one end
- A cartoon character that resembles an actual human and talks like one
- An AI-generated human, that not only looks and talks like an actual human but also mimics actual facial emotions.
If any of these three types compete for the level of an actual human, then that is because of a great execution of an ultra-high-quality, real-time rendered model that would bring life to the data. The approach for creating such kind of bot can be a lot of manual and scripted animation (incl motion capture) and months of character artistry Or it can be a single Deep Learning model, that does the same thing in a fraction of a second.
Deepfakes or Deepbots?
Almost everyone who surfs on the internet has read about or seen Deepfakes. The word Deepfake originates from the terms “deep learning” and “fake”. It is a video or sound synthesis technique based on a specific branch of artificial intelligence. As soon as we hear the term “Deepfakes”, the general attitude tends more towards hostility rather than optimism. But if somehow, we minimize the “fake” component and introduce it into the legislative field, Deepfakes technology can be transformed into Deepbots, which have a number of applications where it can benefit humans on a daily basis.
Synthetic Media Humans (Deepbots)
When an AI model creates a digital human avatar, that interacts and moves like an actual human or like a fictitious human being, we can call it synthetic media humans or Deepbots. Synthetic media humans (Deepbots) is a fairly new branch of the media market, that has attracted a lot of controversy from its onset. On the one hand, synthetic media speeds up content production by making it scalable and affordable, as well as it leverages the power of artificial intelligence to implement ideas that were previously considered impossible. On the other hand, the synthetic media market is still struggling to define the ethical boundaries of its use cases. In simple words, the dilemma is, should this technology be democratized and made available for the masses, or should it be only restricted at the institutional level.
Anyhow, Deepbots or Synthetic Media Humans are quickly transforming the media landscape by creating a positive impact for content creators whose focus is scalability and innovation. Before we discuss some hidden insights of synthetic media humans, let’s first see some common outcomes of these Deepbots.
Outcomes of Deepbots
The absolute nature of technology is never good or bad; the way we use the technology decides its nature. Some good outcomes of Deepbot technology are operational scalability, knowledge scalability, affordability, and visual communication. In simple words, if we take examples:
- Operational scalability: We can reanimate a brand ambassador to give a personalization touch to the advertisement operations of a brand.
- Knowledge scalability: We can create interactable virtual human experts that can hold the same knowledge as a human expert.
- Affordability: We can create a production-grade video of a model talking about your product without worrying about the cameras, sound equipment, or presentation skills of the model.
- Visual communication: We can leverage the visual model of this communication channel (video) to visually convey ideas to our target audience. Research published in the journal of psychological research in cyberspace in 2013, showcased that humans tend to develop a much stronger emotional connection during a video interaction as compared to the text-based chat.
- Multilingual communication: We can make a Deepbot speak many languages and accents for the same text. This helps in breaking the language barrier.
However, keeping aside the ethical side of Deepfakes i.e. Deepbots, even if we use Deepfakes as it is, there are multiple positive outcomes for it as well. For example, imagine yourself as a face in the new Avengers movie, or imagine yourself as hosting the next prime time show. Many people around the world hold these aspirations, and if using technology can fulfill these desires for some, the outcome of tech becomes more fruitful than harmful. A recent controversial example of this is some works of Bill Posters, who follows a tradition of critical public art — like Deepfake Bezos to spread an environmental awareness message for burning forests. This shows that how Deepfakes can be used to spread computational forms of propaganda and exploit the associated architectures that define the Digital Influence Industry.
Deepbots as NextGen Chatbots
Deepbots are virtual reanimation of actual or fictitious humans, that are built to take one side in digital interaction. This is one of the most governable and ethical use cases of this tech.
Presently we can find multiple startups across the world that are building upon this use case. One such example is person8.ai that provides video chatbot solutions to automate customer support and increase engagement for websites. There are many other startups that are focusing on the same line by providing Deepbot video development solutions. One common thing in all of these startups is that they operate at the B2B level, which shows us, how far are we from seeing this technology being democratized.
However, there are some interesting patterns to observe in these developments:
- Generative Artificial Intelligence technology is becoming mainstream. It opens up a whole new micro-industry of synthetic media that holds an enormous capability of disrupting some other linked industries.
- Disruptors are getting disrupted. We saw how chatbots became mainstream within the last few years. Similarly, we would also see video chatbots becoming mainstream in record time.
- Next-generation of user experience. Product builders around the world are always looking to take their features to a point where their products remain a critical step ahead in a competitive market. This comes by either improving your ad-hoc service system or your user experience. This is where the video chatbot solution fits best.
- The shift towards video data. Over the past few decades, we saw a gradual shift from textual data to image data on the internet. But at present, an average social media user sees more minutes of video content in a day than images or text.
Chatbots needs evolution
The biggest barrier to customers using chatbots is that many would prefer to interact with an actual real-life assistant, rather than a machine. But due to a limitation on the human expertise scalability, this is not entirely possible. In the 2018 State of Chatbots Report, about 43% of respondents stopped interacting with chatbots because they wanted to deal with a real-life human on the chat window.
Therefore, we land upon two important questions:
- How can we fulfill the ever-growing need for limitless human interaction?
- How can we promote the culture of self-help, and what is needed to do that?
We can find a solution at the intersection of these two questions, i.e if we can create a system that fills in the place of an actual human (interacts like a real human), as well as if that system can contain and show all possible scenarios to the users visually as per there support request. In this way, we can satisfy more users than before. This is where the video chatbot solutions come in. They are an evolved version of the present text-based chatbots that can visually engage users in a more immersive manner, such that there is a touch of compassion and empathy in the interactions. Video chatbots not only satisfy this ever-rising expectation for a better customer experience but also provide the product owners a way to express their ideas to their users visually.
To back my statement that chatbots do need an evolution, we can see the data of the 2018 State of Chatbots Report which says that about 62% of the baby boomers (older generation) and about 55% of the millennials (younger generation) didn’t like the chatbot experience at all. This not only sets up a low benchmark for such important metrics as customer experience but also shows a large void in the customer support and experience industry.
Can Deepbots based Video Chatbots win users’ Trust?
Users' trust cannot be quantified in a metric, but it can be measured by monitoring their consistency in using a feature, each time they land on your website. A user’s trust can be evaluated in multiple ways just as we evaluate trust in our daily life. You trust your bank to keep your money safe, you trust your doctor’s advice regarding your health and your dog to guard your house. Trust can either be Cognitive or Affective. Cognitive trust is mainly based on the belief in competence, dependability, and reliability — like your trust in your bank. Whereas affective trust is based on the belief in emotional bonds and relationships — like your trust for your family and dog. Similarly, in the case of a user, cognitive and affective trust between your customer support system and your user is based on the interpersonal care between parties. If you are a non-signed-up user, the trust is limited at a cognitive level, but as soon as a user pays for your plan, the trust simply translates its boundaries from the cognitive level to include the affective level.
In the case of video chatbots, where synthetic media humans (Deepbots) is an ingrained part of the interaction, should users trust them in the same way we trust humans? To be honest, we can not trust them exactly as we would trust humans, but we can trust these systems even more. We often see or hear cases, how customer support interactions can be rude, cold, and patience-testing. But in the case of a digital human, there is zero chance of any unpleasant experience happening. My opinion can be easily challenged on multiple grounds depending on the context of customer support we are referring to. Maybe, in many cases, a real human would be of more help to a user than a digital human, but a human also follows a process to provide customer support to a user with a scope of error. Using a digital human would just reduce this error margin even further. Therefore it’s safe to say that eventually, people will build this level of deep trust with AI-based digital humans. However, we still need to be naturally careful as we step into this exciting new world assisted by digital humans.
Should digital human interaction have emotions?
As Humans, we bring emotional entanglement in all our interactions, no matter if the feelings behind those emotions are irrational, illogical, or personal. Even when you are interacting with a text-based chatbot, there are silent emotions getting generated in our subconscious. If the chatbot is unable to help you, that silent emotion can turn into anger or disappointment in a fraction of a second. However, if you are talking to an actual customer success manager, due to an emotional exchange from both sides, you are less likely to get angry or disappointed. Therefore, emotions are absolutely necessary in a two-way interaction in the case of digital humans too. As a matter of fact, the facial and audio emotions of digital humans are one of the key features that makes them different from text-based bots.
It's a consensus that humans filter their perceived world through a sieve of their own past experience and internal state. If you create a user experience that is emotionally negative for your users (font, website color), they will hardly stay on your website. But if you create an emotionally positive experience that promotes the feeling of security and trust, your users are more likely to feel an incubation of affective trust. Affective trust develops a feeling of personalization and is more productive in the case of user experience. Similarly, when a video chatbot, addresses your user, the facial and audio emotions of the digital human generate authentic intimacy, empathy, and a feeling of friendship for the user. Such feelings catalyze powerful momentum value propositions and customer retention for your brand.
Final Thoughts
The Digital Human wave is a growing movement that is shaping the future of the internet and especially the content we digest on websites. If you want to check some digital humans in interactions, some examples are:
- Lil Miquela, a digital influencer based on digital human
- Video Chatbot based on Digital human by person8.ai
- McDonald’s digital human cashier by Soul Machines
There are many more digital humans that are available on the internet. These digital humans are eventually emerging on social media and private communication channels, as internet users are starting to relate to these bots on a human level and project an understanding of how these bots can help them in cases where would be left unattended otherwise. Digital humans are not only designed for embodied and meaningful interactions but they are also designed by embracing the principles that fabricate the building blocks of trust in society. Users’ trust for digital human experience will only grow if the users know the origin of this technology and the system behind them.
In the gusts of this growing online world, digital trust is becoming a global commodity that is difficult to acquire and quick to lose. If we keep in mind the human need for connection at the heart while developing and designing technology, we can build and maintain trust over time. As a result of this trust, we can leverage the power of these synthetic media humans to create digital solutions that can help people in ways we’re only just starting to speculate.
My blogs are a reflection of what I worked on and simply convey my understanding of these topics. My interpretation of these topics can be different from that of yours, but my interpretation can only be as inerrant as I am.
