The rise of Virtual Assistants

The rapid rise of artificial intelligence systems for use in virtual assistant technology in recent years underlines how it is increasingly helping people in their everyday lives.

The rapid rise of artificial intelligence (AI) systems for use in virtual assistant (VA) technology in recent years underlines their increasing importance for helping people in their everyday lives.  Driven by more sophisticated computer programming, enormous advances in voice recognition allow VAs to user questions: they are increasingly supporting people in office environments and across various sectors, such as health, auto and home design.

Product structures 

The arrival of VAs brings with it a debate about the direction of product structures: will there be numerous specialised assistants? Or will we have a super assistant capable of conducting a variety of tasks? It is likely the future lies somewhere between the two. It is useful to distinguish between horizontal and vertical VAs in order to frame the discussion about their capabilities: horizontal assistants can help with a variety of precise tasks, yet find it difficult to carry out more complex and multidimensional assignments; vertical assistants, however, are more likely to complete larger end-to-end tasks that may take longer to accomplish.

While VAs are not yet able to accomplish all the specialised tasks that require an iterative process with the user, in the future we may see basic standard platforms develop. This could pave the way for a common AI-VA platform allowing for specialist VAs. 

the entertaining side of virtual assistants
Tars, the live-wire robot in the film ‘Interstellar’
Source: L’Atelier
Apple’s virtual assistant Siri has been known to be able to make sarcastic remarks
Source: Gizmodo
Théodore, hero of the film Her, falls in love with his operating system
Source: Wild Bunch distribuition

<!–

The entertaining side of virtual assistants

Tars, the live-wire robot in the film “Interstellar”
Apple’s virtual assistant Siri has been known to be able to make sarcastic remarks
Théodore, hero of the film “Her”, falls in love with his operating system
–>

The potential for “deep learning”

Evolution, adaptability and the capacity to react to the unknown are essential components of what it means to be human. “Deep learning”, where machines can participate in complex discussions by picking responses from a large database, relies on these characteristics, but is not yet capable of replicating them. A programme capable of adapting on its own and solving problems for which it has not been prepared – in other words, unsupervised learning – has tremendous potential. It will be capable of writing its own programme, allowing it to carry out a potentially infinite number of actions. With this potential, care needs to be taken to ensure that the technology developed in Silicon Valley does not impose its norms on the rest of the world: technology exists to serve people.

The future of VAs

Indications show a shift in the VA market towards an oligopolistic structure, where a few large technology giants buy smaller specialised start-ups.  Existing internet leaders appear to have embarked on a pitched battle to acquire inventive young start-ups to help improve their products and user experience.

In the long term, the creation of a platform designed to be integrated into connected objects could enable the interaction of various devices to enhance user experience.  Incumbents would need to adapt their business models to the new environment. For example in the automotive sector, the value of a vehicle may gravitate slowly away from the hardware to the software, as technology that integrates VAs into vehicles will become more common. Additionally banks may begin to incorporate VAs into their investment offering for customers, in line with a client’s interes


Rudi Collin (Global Head of Digital Transformation at BNP Paribas CIB) understands the impact of these technological advances: