Satya Nadella, CEO of Microsoft used to says that “Bots are new apps”, referring to the boom and potential of the bot ecosystem, similar to what happened in the ecosystem of applications.

Another giant- Facebook has advanced even more in the field: Facebook Messenger has supported 100,000 developers in just over a year and achieved 100,000 bots ( comparing to 34,000 developers and 30,000 bots in 2016). Obviously, not only is Microsoft CEO making this statement, but the “Big 4″ leaders in tech circles like Google, Facebook, Apple, and Microsoft also share the same viewpoint when they have made progress in building artificial intelligence its own bot ecosystem through building platform, or tools to create bot …(More details in the next section:”Bot ecosystem in the world – looming battlefield “)

Gartner Hype Cycle – Emerging Trend in 2016 (Source: Gartner)

“Conversational User interfaces”, which is bot & chatbot, “Virtual Personal Assistants,” is in the “innovation triggering” phase and has growth potential. The initial power when large companies invest heavily in this field. Meanwhile, “Machine Learning” is reaching its peak in the “Peak of Inflated Expectation”, the field of “Natural-Language Question Answering” starts to go down, at the stage “Trough of Disillusionment” but soon reach “slope of enlightment”

Table 1. Gartner Hype Cycle

The Term ‘Hype Cycle’ is used to refer to a graphical representation of the stages of the life cycle of technology that starts from its birth to its maturity and finally to its widespread usage and adoption

Technology trigger – Technology Trigger is the first stage in the emergence of a new technology and is a stage in which a potential technology gets a breakthrough or kicks off.

Peak of inflated expectations – This is the next stage in the Hype Cycle and is a stage that is associated with over-enthusiasm about the new technology. This over-enthusiasm is created by the company with the help of media channels like newspapers, ads, magazines, social media, etc.

“Trough of disillusionment”- The technology now becomes unfashionable, and no one is talking about it anymore. People are bored of talking about the technology, and even the creators see no point in publicizing it anymore.

Slope of enlightenment –This is the stage when people begin to understand how the technology can prove beneficial or useful. This is often the result of hard work as well as focused experimentation by some organizations.

Plateau of productivity – Now that the main benefits and practical application of the technology have become wide-spread knowledge, mainstream adoption begins to take place.

So what is underneath these waves? There are three main trends:

The first trend account for the rise of bot is booming of Artificial Intelligence (AI). AI applications include virtual assistants (including chatbot), system recommendations, voice analysis, and more (see more in the table below)

Main areas of AI (Source: PwC)

  • Large-scale Machine Learning: Design of learning algorithms, as well as scaling existing algorithms, to work with extremely large data sets.
  • Deep Learning: Model composed of inputs such as image or audio and several hidden layers of sub-models that serve as input for the next layer and ultimately an output or activation function.
  • Natural Language Processing (NLP): Algorithms that process human language input and convert it into understandable representations.
  • Collaborative Systems: Models and algorithms to help develop autonomous systems that can work collaboratively with other systems and with humans.
  • Computer Vision (Image Analytics): The process of pulling relevant information from an image or sets of images for advanced classification and analysis.
  •  Soft Robotics (Robotic Process Automation): Automation of repetitive tasks and common processes such as IT, customer servicing and sales without the need to transform existing IT system maps.

The level of interest and investment for AI is extremely high and is being demonstrated by the numbers: According to McKinsey, companies have invested in AI from $ 26 billion to $ 39 billion in 2016. And according to Accenture report, CAGR, which is the annual cumulative annual growth rate is 50% since 2010. The reason AI has become a trend in the past few years is due to the increased data volume powerful processing capability as well as storage capacity, reduced cost of processing power.

3 capabilities of AI (Source: Accenture)

The branch of AI: Sense (with technologies such as computer vision, audio processing …), Comprehend  (with technologies such as natural language processing – natural language processing, …), or Act (machine-learning) achieved incredible milestone. For example, when using deep learning, image recognition can reduce errors by 25% to 2%, far more than human capacity.

The second trend is the fatigue of users due to more and more applications jamming in smartphones. In fact, the number of smartphones is growing well with so many models, and the number of downloads of applications is also increasing, but app downloads are reaching consolidation.

It seems that the ease of doing business on the content store has been a thing of the past, and building an app and promoting it is increasingly costly. There is a good observation of this situation by the Economist: “The 20 most successful developers grab nearly half of all revenues on Apple’s app store”. A quarter of the downloaded applications were deleted after the initial download. Users who have too many choices on smartphones and lack a certain aggregation application that manage functions like other applications.

The third trend is that users are “tired” but still are increasingly converge to spend more time on messaging applications (according to Deloitte, the number of monthly users. Big 4 in messaging platform up to 3 billion – including Facebook, Whatsapp, Line, Wechat). There are more than 2.5 billion people who have at least one chat message installed in the machine. This number will rapidly increase to 3.6 billion people – half our humanity in the coming years!

Number of users using messaging applications by month (Source: Statista)

Even teenagers tend to use the messaging app more than social networks. We should take into account that the top four messaging platforms outperformed the social networking platform in terms of usage (according to BI Intelligence). This also explains why Facebook emphasize mobile & splited the two social networking and messaging applications platform.

So why the messaging app can not be a “master” application as mentioned above? The fact is that the main function of the phone is receiving, giving call , messaging, so why not put more functionality into an application instead of moving from one application to the other? Messenger app for a large share of the user’s day-to-day budget and is becoming a “master” application, an “operating system” of applications that can replace the functionality of other applications when combining chatbot on this platform.

Social network app vs Messenger app (Source: BI Intelligence)

A very specific example in Asia is Wechat, Tencent’s messaging platform not only does well messaging and conversation but also extends to many other areas such as doctor appointments, booking meals o, or even giving digital lucky money in the Lunar New Year. Not only Wechat but other messaging platforms around the world are also geared towards this “master” model.

Obviously, what user wants is the convenience in experience, optimization of time & cost. Consequently, the great potential of the “bot” is undeniable. However, whether or not bot ecosystem having reached the level of development like in application ecosystem is still an open question. In the next section, the use cases of bot, bot ecosystems in the world and in Vietnam will be explored further.

Table 2. Bot concept  chatbot development & history

Gartner defines the term “bot” as relatively new, but the technology that makes the bot has evolved before. Bot is a small piece of code or an application that performs a specialized task. Bots will use AI to manage unstructured data and complex tasks.

Chatbots, a specific type of bot that fits human interaction, uses AI to handle language. Chatbot can simply understand a computer program that a user can communicate with the machine through the messaging application. A chatbot can speak and understand speech and will analyze what people say and try to understand a given request. Chatbots then communicate with other machines and convey the question and finally give feedback to human.

Bot, or chatbot, helps people save time, costs through application in customer care (Process automation …), or improve productivity (bots help to schedule things, optimize cost …) or even taking care of human life (health care bots …).

Although chatbot has been a hot topic lately, chatbot has been around for 50 years. In 1950, Turing’s idea was to offer a clever device that would replace human beings with conversations. This idea forms the basis of the Chatbot revolution. Then, Eliza was the first chatbot program developed in 1966. The program was created to take part as the therapist to answer simple simple questions with crystal clear sentence structures. The program was developed by Joseph Weizenbaum, Massachusetts Institute of Technology, USA.

Hoang Nam Le

Related posts: